From e7cb71c64250cc107fb864a878226ce3ec963923 Mon Sep 17 00:00:00 2001
From: Radek Czemerys <7029942+radko93@users.noreply.github.com>
Date: Tue, 14 Apr 2026 20:37:17 +0800
Subject: [PATCH 1/5] fix: graceful degradation when native libraries are
 unavailable (Android) (#1067)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

## Description

On 32-bit Android devices (or devices with 64-bit CPUs running 32-bit
userspace), the `arm64-v8a` native `.so` files are not found.
`ETInstaller.init` throws `RuntimeException` wrapping
`UnsatisfiedLinkError`, crashing the app at startup with no
consumer-side escape hatch.

This PR adds a fallback TurboModule (`ETInstallerUnavailable`) whose
`install()` returns `false`. `RnExecutorchPackage` returns it only when
the native library fails to load, so JS sees a real linked module but
JSI bindings are never injected. This preserves the existing
linking-error Proxy for genuinely mislinked installs on supported
devices.

**Unsupported ABI**: real module exists → `install()` returns `false` →
globals not set → `isAvailable` is `false`
**Supported but mislinked**: module is `null` → existing Proxy throws
linking error (unchanged)

No public API break; preserves existing mislink failure behavior.
`isAvailable` is a new additive export.

### Introduces a breaking change?

- [x] No

### Type of change

- [x] Bug fix (change which fixes an issue)

### Tested on

- [x] Android

### Testing instructions

1. Build and run on a 32-bit Android device (e.g. Galaxy A13, Moto G
Play 2023)
2. App should start without crashing
3. `isAvailable` should be `false`
4. Verify on a supported 64-bit device that everything works as before

### Related issues

Fixes #1065

(cherry picked from commit 8f20cee6e57447c6f15fd9184a00fec5995ad7e1)
---
 .../rnexecutorch/ETInstallerUnavailable.kt    | 27 +++++++++++++++++++
 .../rnexecutorch/RnExecutorchPackage.kt       | 13 ++++++++-
 packages/react-native-executorch/src/index.ts | 11 ++++++++
 3 files changed, 50 insertions(+), 1 deletion(-)
 create mode 100644 packages/react-native-executorch/android/src/main/java/com/swmansion/rnexecutorch/ETInstallerUnavailable.kt

diff --git a/packages/react-native-executorch/android/src/main/java/com/swmansion/rnexecutorch/ETInstallerUnavailable.kt b/packages/react-native-executorch/android/src/main/java/com/swmansion/rnexecutorch/ETInstallerUnavailable.kt
new file mode 100644
index 0000000000..0b4c57bb60
--- /dev/null
+++ b/packages/react-native-executorch/android/src/main/java/com/swmansion/rnexecutorch/ETInstallerUnavailable.kt
@@ -0,0 +1,27 @@
+package com.swmansion.rnexecutorch
+
+import com.facebook.react.bridge.ReactApplicationContext
+import com.facebook.react.bridge.ReactMethod
+import com.facebook.react.common.annotations.FrameworkAPI
+import com.facebook.react.module.annotations.ReactModule
+
+/**
+ * Fallback TurboModule returned when native ExecuTorch libraries cannot be
+ * loaded (e.g. 32-bit Android devices where only arm64-v8a binaries are
+ * shipped). Extends the same spec as ETInstaller so JS sees a real linked
+ * module, but install() returns false to signal unavailability.
+ */
+@OptIn(FrameworkAPI::class)
+@ReactModule(name = ETInstallerUnavailable.NAME)
+class ETInstallerUnavailable(
+  reactContext: ReactApplicationContext,
+) : NativeETInstallerSpec(reactContext) {
+  companion object {
+    const val NAME = NativeETInstallerSpec.NAME
+  }
+
+  @ReactMethod(isBlockingSynchronousMethod = true)
+  override fun install(): Boolean {
+    return false
+  }
+}
diff --git a/packages/react-native-executorch/android/src/main/java/com/swmansion/rnexecutorch/RnExecutorchPackage.kt b/packages/react-native-executorch/android/src/main/java/com/swmansion/rnexecutorch/RnExecutorchPackage.kt
index 0b15e216a5..46abca85d7 100644
--- a/packages/react-native-executorch/android/src/main/java/com/swmansion/rnexecutorch/RnExecutorchPackage.kt
+++ b/packages/react-native-executorch/android/src/main/java/com/swmansion/rnexecutorch/RnExecutorchPackage.kt
@@ -15,7 +15,18 @@ class RnExecutorchPackage : TurboReactPackage() {
     reactContext: ReactApplicationContext,
   ): NativeModule? =
     if (name == ETInstaller.NAME) {
-      ETInstaller(reactContext)
+      try {
+        ETInstaller(reactContext)
+      } catch (e: RuntimeException) {
+        if (e.cause is UnsatisfiedLinkError) {
+          // Native library not available (e.g. 32-bit device without arm64-v8a .so).
+          // Return a fallback module whose install() returns false so JS can
+          // distinguish "unsupported ABI" from "package not linked."
+          ETInstallerUnavailable(reactContext)
+        } else {
+          throw e
+        }
+      }
     } else {
       null
     }
diff --git a/packages/react-native-executorch/src/index.ts b/packages/react-native-executorch/src/index.ts
index 3f6b9fc4ce..afd90bc780 100644
--- a/packages/react-native-executorch/src/index.ts
+++ b/packages/react-native-executorch/src/index.ts
@@ -134,9 +134,20 @@ if (
       `Failed to install react-native-executorch: The native module could not be found.`
     );
   }
+  // install() returns false when the native library is intentionally unavailable
+  // (e.g. ETInstallerUnavailable on unsupported ABIs). In that case, JSI
+  // bindings are not injected and globals remain unset.
   ETInstallerNativeModule.install();
 }
 
+/**
+ * Whether the native ExecuTorch runtime is available on this device.
+ * Returns `false` when native libraries cannot be loaded (e.g. 32-bit Android
+ * devices where only arm64-v8a binaries are shipped).
+ * @category Utilities - General
+ */
+export const isAvailable = typeof global.loadExecutorchModule === 'function';
+
 // hooks
 export * from './hooks/computer_vision/useClassification';
 export * from './hooks/computer_vision/useObjectDetection';

From ee8d0a20177b643b8162e4658ede26c7f02b50d3 Mon Sep 17 00:00:00 2001
From: Mateusz Sluszniak <56299341+msluszniak@users.noreply.github.com>
Date: Wed, 22 Apr 2026 17:17:55 +0200
Subject: [PATCH 2/5] fix(llm): normalize multimodal image paths to file://
 URIs (#1090)

## Description

`LLMController.forward` passed `imagePaths` straight through to
`nativeModule.generateMultimodal` with no normalization. The native side
requires the `file://` prefix; without it, native throws `"Read image
error: invalid argument"` with no further context. Callers can plausibly
arrive with either form:

- `ResourceFetcher.fetch` returns raw paths *without* `file://` (per its
own docstring on the `fetch` method).
- Platform image-picker APIs (e.g. `expo-image-picker`) typically return
`file:///...` URIs.
- The same path string passed to a vision module's `forward(...)` works
either way; the asymmetry between vision modules and multimodal LLM is
undocumented.

This PR normalizes each image path inside `LLMController.forward` so
both forms work, and updates the JSDoc on `Message.mediaPath` and
`LLMModule.forward.imagePaths` to document the new contract.

### Introduces a breaking change?

- [ ] Yes
- [x] No

Strictly additive: previously-working calls (paths with `file://`) keep
working unchanged. Previously-failing calls (paths without `file://`)
now succeed.

### Type of change

- [x] Bug fix (change which fixes an issue)
- [ ] New feature
- [ ] Documentation update
- [ ] Other

### Tested on

- [ ] iOS
- [ ] Android

The bare-path failure was reproduced on Android (Samsung Galaxy S24
Ultra) with LFM2-VL-1.6B while building a downstream consumer; both
forms tested manually post-fix on the same device. Re-verification of
both forms on iOS is recommended.

### Testing instructions

```ts
import { LLMModule, LFM2_VL_1_6B_QUANTIZED } from 'react-native-executorch';
const llm = await LLMModule.fromModelName(LFM2_VL_1_6B_QUANTIZED);

// Both should now work; previously only the first did.
await llm.generate([
  { role: 'user', content: 'Describe.', mediaPath: 'file:///absolute/path/to/img.jpg' },
]);
await llm.generate([
  { role: 'user', content: 'Describe.', mediaPath: '/absolute/path/to/img.jpg' },
]);
```

### Related issues

Addresses item 3 of #1086.

### Checklist

- [ ] I have performed a self-review of my code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have updated the documentation accordingly
- [ ] My changes generate no new warnings

### Additional notes

The normalizer is module-scope (matching `messagesForChatTemplate` from
#1089) rather than a class method because it doesn't depend on
controller state.

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
(cherry picked from commit e8d4305ee42a002f634fa12076b7a1b4512c4a21)
---
 .../src/controllers/LLMController.ts               | 14 +++++++++++++-
 .../natural_language_processing/LLMModule.ts       |  2 +-
 packages/react-native-executorch/src/types/llm.ts  |  2 ++
 3 files changed, 16 insertions(+), 2 deletions(-)

diff --git a/packages/react-native-executorch/src/controllers/LLMController.ts b/packages/react-native-executorch/src/controllers/LLMController.ts
index 7ce921f6f5..8829f37106 100644
--- a/packages/react-native-executorch/src/controllers/LLMController.ts
+++ b/packages/react-native-executorch/src/controllers/LLMController.ts
@@ -254,7 +254,7 @@ export class LLMController {
         imagePaths && imagePaths.length > 0
           ? await this.nativeModule.generateMultimodal(
               input,
-              imagePaths,
+              imagePaths.map(normalizeImagePath),
               this.getImageToken(),
               this.onToken
             )
@@ -456,3 +456,15 @@ export class LLMController {
     return result;
   }
 }
+
+/**
+ * The native multimodal pipeline expects image paths to be `file://` URIs.
+ * `ResourceFetcher.fetch` and most platform file APIs return raw filesystem
+ * paths without that prefix, so callers routinely pass either form. Accept
+ * both and normalize to the prefixed form here.
+ * @param path - Local image path, either with or without the `file://` prefix.
+ * @returns The same path with a `file://` prefix.
+ */
+function normalizeImagePath(path: string): string {
+  return path.startsWith('file://') ? path : `file://${path}`;
+}
diff --git a/packages/react-native-executorch/src/modules/natural_language_processing/LLMModule.ts b/packages/react-native-executorch/src/modules/natural_language_processing/LLMModule.ts
index f7648a7b95..bdb5ada699 100644
--- a/packages/react-native-executorch/src/modules/natural_language_processing/LLMModule.ts
+++ b/packages/react-native-executorch/src/modules/natural_language_processing/LLMModule.ts
@@ -139,7 +139,7 @@ export class LLMModule {
    * It doesn't manage conversation context. It is intended for users that need access to the model itself without any wrapper.
    * If you want a simple chat with model the consider using `sendMessage`
    * @param input - Raw input string containing the prompt and conversation history.
-   * @param imagePaths - Optional array of local image paths for multimodal inference.
+   * @param imagePaths - Optional array of local image paths for multimodal inference. Each entry may be either `file:///absolute/path` or `/absolute/path` — the controller normalizes the path before passing it to native code.
    * @returns The generated response as a string.
    */
   async forward(input: string, imagePaths?: string[]): Promise<string> {
diff --git a/packages/react-native-executorch/src/types/llm.ts b/packages/react-native-executorch/src/types/llm.ts
index 83f9630c10..3dbf54bb94 100644
--- a/packages/react-native-executorch/src/types/llm.ts
+++ b/packages/react-native-executorch/src/types/llm.ts
@@ -270,6 +270,8 @@ export interface Message {
   /**
    * Optional local file path to media (image, audio, etc.).
    * Only valid on `user` messages.
+   * Either `file:///absolute/path` or `/absolute/path` is accepted; the
+   * controller normalizes the path before passing it to native code.
    */
   mediaPath?: string;
 }

From 7d8d8f19e9e961daf8568d2d9399d414413f16ba Mon Sep 17 00:00:00 2001
From: Mateusz Sluszniak <56299341+msluszniak@users.noreply.github.com>
Date: Wed, 22 Apr 2026 18:08:28 +0200
Subject: [PATCH 3/5] fix(llm): auto-shape multimodal mediaPath messages in
 chat template (#1089)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

## Description

`LLMController.generate()` collected `imagePaths` from messages with a
`mediaPath` set, but never transformed their `content` into the
`[{type:'image'}, {type:'text', text}]` form that the chat template
needs to emit the `<image>` placeholder. Calling `generate()` directly
with a vision-capable model (e.g. LFM2-VL) thus threw `"More images
paths provided than '<image>' placeholders in prompt"` from native, even
though `sendMessage()` worked because it built its own
`historyForTemplate` that did the transformation.

This PR moves the transformation into `applyChatTemplate` so both call
sites (`generate` and `sendMessage`) get the correct behavior, and
removes the now-redundant `historyForTemplate` block from `sendMessage`.
The public `Message.content` type stays `string` — external callers
always pass plain strings; the controller handles the structured array
form internally.

The helper is idempotent: messages whose `content` is already an array
(e.g. callers who pre-shaped it as a workaround) are passed through
unchanged.

### Introduces a breaking change?

- [ ] Yes
- [x] No

Public types are unchanged. `sendMessage` produces an identical rendered
chat-template string (the transformation just happens one step later in
the pipeline; token count and rendered output are byte-identical).
`generate` only changes behavior in cases that previously threw — pure
bug fix.

### Type of change

- [x] Bug fix (change which fixes an issue)
- [ ] New feature (change which adds functionality)
- [ ] Documentation update (improves or adds clarity to existing
documentation)
- [ ] Other (chores, tests, code style improvements etc.)

### Tested on

- [ ] iOS
- [ ] Android

The original bug was reproduced on a vision-capable model
(LFM2-VL-1.6B-quantized) on Android while building a downstream consumer
app. Re-verification of the fix on a real device is recommended before
merge — see Testing instructions below. ~I have not personally re-run
the failing scenario after the fix.~

### Testing instructions

To reproduce the original bug (without this PR):

```ts
import { LLMModule, LFM2_VL_1_6B_QUANTIZED } from 'react-native-executorch';

const llm = await LLMModule.fromModelName(LFM2_VL_1_6B_QUANTIZED);
await llm.generate([
  { role: 'user', content: 'Describe this image.', mediaPath: 'file:///path/to/image.jpg' },
]);
// Throws: "More images paths provided than '<image>' placeholders in prompt"
```

With this PR applied, the same call should succeed and return the
model's description.

Regression check: a vision-capable `sendMessage(text, { imagePath })`
flow should continue producing identical output.

### Screenshots

N/A (controller change, no UI).

### Related issues

Addresses items 1 and 2 of #1086. With item 1 fixed, item 2's
`Message.content` type mismatch no longer surfaces in practice because
external callers never need to construct the array form themselves (the
`as unknown as string` workaround that motivated #2 becomes
unnecessary).

### Checklist

- [ ] I have performed a self-review of my code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have updated the documentation accordingly
- [ ] My changes generate no new warnings

### Additional notes

The `messagesForChatTemplate` helper lives at module scope rather than
as a static class method because it doesn't depend on controller state.
Internal `any[]` return is a deliberate concession to the dynamic shape
the chat-template engine accepts; the public `Message[]` input/output
contract stays well-typed.

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
(cherry picked from commit b271aa61d8d7c039fcbb2e1cf8bad70f026a0904)
---
 .../src/controllers/LLMController.ts          | 37 ++++++++++++-------
 1 file changed, 23 insertions(+), 14 deletions(-)

diff --git a/packages/react-native-executorch/src/controllers/LLMController.ts b/packages/react-native-executorch/src/controllers/LLMController.ts
index 8829f37106..d9d5593ae3 100644
--- a/packages/react-native-executorch/src/controllers/LLMController.ts
+++ b/packages/react-native-executorch/src/controllers/LLMController.ts
@@ -354,18 +354,6 @@ export class LLMController {
     const updatedHistory = [...this._messageHistory, newMessage];
     this.messageHistoryCallback(updatedHistory);
 
-    const historyForTemplate = updatedHistory.map((m) =>
-      m.mediaPath
-        ? {
-            ...m,
-            content: [
-              { type: 'image' },
-              { type: 'text', text: m.content },
-            ] as any,
-          }
-        : m
-    );
-
     const visualTokenCount = this.nativeModule.getVisualTokenCount();
     const countTokensCallback = (messages: Message[]) => {
       const rendered = this.applyChatTemplate(
@@ -383,7 +371,7 @@ export class LLMController {
     const messageHistoryWithPrompt =
       this.chatConfig.contextStrategy.buildContext(
         this.chatConfig.systemPrompt,
-        historyForTemplate,
+        updatedHistory,
         maxContextLength,
         countTokensCallback
       );
@@ -448,7 +436,7 @@ export class LLMController {
     );
 
     const result = template.render({
-      messages,
+      messages: messagesForChatTemplate(messages),
       tools,
       ...templateFlags,
       ...specialTokens,
@@ -468,3 +456,24 @@ export class LLMController {
 function normalizeImagePath(path: string): string {
   return path.startsWith('file://') ? path : `file://${path}`;
 }
+
+/**
+ * Multimodal chat templates expect message content for image-bearing turns
+ * to be an array of content parts with an `image` part as a placeholder.
+ * Callers of `LLMController.generate` and `LLMController.sendMessage` pass
+ * messages with a plain string `content` plus an optional `mediaPath`; this
+ * helper rewrites them into the structured form that the template engine
+ * understands.
+ * @param messages - Messages to prepare for the chat template engine.
+ * @returns Messages with image-bearing turns rewritten to structured content.
+ */
+function messagesForChatTemplate(messages: Message[]): any[] {
+  return messages.map((m) =>
+    m.mediaPath && typeof m.content === 'string'
+      ? {
+          ...m,
+          content: [{ type: 'image' }, { type: 'text', text: m.content }],
+        }
+      : m
+  );
+}

From 8b4eee15253148dcd978e7e1ecfa6bbde4af7f17 Mon Sep 17 00:00:00 2001
From: Norbert Klockiewicz <Nklockiewicz12@gmail.com>
Date: Tue, 28 Apr 2026 15:26:46 +0200
Subject: [PATCH 4/5] feat(llm): min_p and repetition_penalty sampling,
 per-model defaults, letterbox vision (#1099)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

## Description

Adds `min_p` and `repetition_penalty` sampling parameters to
`GenerationConfig`, plumbs them through the full stack (`Sampler` →
`TextDecoderRunner` → `TextTokenGenerator` → `BaseLLMRunner` /
`TextRunner` / `MultimodalRunner` → JSI bindings → `LLMController`),
introduces a per-model default `generationConfig` that gets applied
automatically on load (populated for Qwen3 and LFM2-VL from their
upstream recommendations), and replaces the distorting `cv::resize` in
`VisionEncoder` with the existing `resizePadded` helper so multimodal
inputs keep their aspect ratio. Also fixes three silent pre-existing
bugs surfaced along the way: an xorshift PRNG seeded with `0` that made
sampling deterministic, a `Sampler::apply_min_p` renormalization gap,
and inline `{}` no-op overrides in `MultimodalRunner` that would desync
in future refactors.

### Introduces a breaking change?

- [ ] Yes
- [x] No

### Type of change

- [x] Bug fix (change which fixes an issue)
- [x] New feature (change which adds functionality)
- [ ] Documentation update (improves or adds clarity to existing
documentation)
- [ ] Other (chores, tests, code style improvements etc.)

### Tested on

- [x] iOS
- [ ] Android

### Testing instructions

**Sampling parameter plumbing**
1. Open `apps/llm`, load any supported model (e.g.
`LFM2_VL_450M_QUANTIZED`).
2. Without any manual `configure()` call, send a prompt. The model card
defaults are applied automatically — for LFM2-VL you should now see
coherent, non-repetitive descriptions (previously the model often
produced generic or looping replies at the library's default
`temperature=0.8, topp=0.9`).
3. Optionally override via `useLLM(...)`'s `configure({
generationConfig: { temperature: 0.7, minP: 0.1, repetitionPenalty: 1.05
} })` and confirm the generation style changes.

**Letterbox preprocessing**
1. With a multimodal model loaded in `apps/llm` → `multimodal_llm`
screen, attach a photo with a non-square aspect ratio (e.g. 3000×2250
from your camera roll).
2. Ask the model to describe it. Before this PR the image was stretched
into the PTE's square input shape — the model would sometimes
misidentify subjects in wide/tall photos. After, the image is
letterboxed so proportions are preserved.

### Screenshots

<!-- none -->

### Related issues

<!-- none -->

### Checklist

- [x] I have performed a self-review of my code
- [x] I have commented my code, particularly in hard-to-understand areas
- [x] I have updated the documentation accordingly
- [x] My changes generate no new warnings

### Additional notes

**Per-model recommended defaults**

Model presets gain an optional `generationConfig` field;
`LLMController.load` applies it before flipping `isReady`, so users see
sensible sampling out of the box. User `configure()` calls still
override per-field. Populated for:

- **Qwen3** family (`temperature=0.6, topp=0.95`, from
`generation_config.json`)
- **LFM2-VL** family (`temperature=0.1, minP=0.15,
repetitionPenalty=1.05`, from the LiquidAI model card)

Other presets (Llama, SmolLM2, Hammer, Phi-4, Qwen2.5, LFM2 text) keep
the library defaults — these model cards don't publish sampling
recommendations, so adding arbitrary values would be guessing.

---------

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
(cherry picked from commit 9f1c89f8228129736a9564529f3918cb63d4ec03)
---
 apps/llm/app/multimodal_llm/index.tsx         |  4 +-
 .../01-natural-language-processing/useLLM.md  | 22 +++--
 .../LLMModule.md                              | 14 ++-
 .../host_objects/ModelHostObject.h            |  9 ++
 .../common/rnexecutorch/models/llm/LLM.cpp    | 24 ++++++
 .../common/rnexecutorch/models/llm/LLM.h      |  2 +
 .../common/rnexecutorch/tests/CMakeLists.txt  |  6 ++
 .../tests/integration/LLMTest.cpp             | 25 ++++++
 .../tests/integration/stubs/StubRunner.h      |  5 --
 .../rnexecutorch/tests/unit/RunnerTest.cpp    | 15 +++-
 .../rnexecutorch/tests/unit/SamplerTest.cpp   | 85 +++++++++++++++++++
 .../common/runner/base_llm_runner.cpp         | 14 +--
 .../common/runner/base_llm_runner.h           | 12 ++-
 .../common/runner/encoders/vision_encoder.cpp | 16 +++-
 .../common/runner/irunner.h                   | 15 ++++
 .../common/runner/multimodal_decoder_runner.h |  5 +-
 .../common/runner/multimodal_runner.cpp       | 20 +----
 .../common/runner/multimodal_runner.h         |  4 -
 .../common/runner/sampler.cpp                 | 45 +++++++---
 .../common/runner/sampler.h                   | 60 ++++++++++++-
 .../common/runner/text_decoder_runner.cpp     | 34 +++++++-
 .../common/runner/text_decoder_runner.h       | 59 +++----------
 .../common/runner/text_runner.cpp             | 29 +------
 .../common/runner/text_runner.h               |  4 -
 .../common/runner/text_token_generator.h      | 38 +++++----
 .../src/constants/modelUrls.ts                | 47 +++++++++-
 .../src/controllers/LLMController.ts          | 58 ++++++++++---
 .../natural_language_processing/useLLM.ts     |  1 +
 .../react-native-executorch/src/types/llm.ts  | 20 ++++-
 29 files changed, 514 insertions(+), 178 deletions(-)
 create mode 100644 packages/react-native-executorch/common/rnexecutorch/tests/unit/SamplerTest.cpp

diff --git a/apps/llm/app/multimodal_llm/index.tsx b/apps/llm/app/multimodal_llm/index.tsx
index 0cbc7d06fc..14a17eff21 100644
--- a/apps/llm/app/multimodal_llm/index.tsx
+++ b/apps/llm/app/multimodal_llm/index.tsx
@@ -14,7 +14,7 @@ import {
 import { launchImageLibrary } from 'react-native-image-picker';
 import { useIsFocused } from '@react-navigation/native';
 import { useSafeAreaInsets } from 'react-native-safe-area-context';
-import { useLLM, LFM2_VL_1_6B_QUANTIZED } from 'react-native-executorch';
+import { useLLM, LFM2_5_VL_1_6B_QUANTIZED } from 'react-native-executorch';
 import SendIcon from '../../assets/icons/send_icon.svg';
 import PauseIcon from '../../assets/icons/pause_icon.svg';
 import ColorPalette from '../../colors';
@@ -50,7 +50,7 @@ function MultimodalLLMScreen() {
   const [error, setError] = useState<string | null>(null);
 
   const vlm = useLLM({
-    model: LFM2_VL_1_6B_QUANTIZED,
+    model: LFM2_5_VL_1_6B_QUANTIZED,
   });
   const tokenCount = vlm.isReady ? vlm.getGeneratedTokenCount() : 0;
   const { stats, onMessageSend } = useLLMStats(
diff --git a/docs/docs/03-hooks/01-natural-language-processing/useLLM.md b/docs/docs/03-hooks/01-natural-language-processing/useLLM.md
index 15d6d3191f..a04f4326bd 100644
--- a/docs/docs/03-hooks/01-natural-language-processing/useLLM.md
+++ b/docs/docs/03-hooks/01-natural-language-processing/useLLM.md
@@ -208,7 +208,15 @@ To configure model (i.e. change system prompt, load initial conversation history
 
   - [`temperature`](../../06-api-reference/interfaces/GenerationConfig.md#temperature) - Scales output logits by the inverse of temperature. Controls the randomness / creativity of text generation.
 
-  - [`topp`](../../06-api-reference/interfaces/GenerationConfig.md#topp) - Only samples from the smallest set of tokens whose cumulative probability exceeds topp.
+  - [`topP`](../../06-api-reference/interfaces/GenerationConfig.md#topp) - Only samples from the smallest set of tokens whose cumulative probability exceeds topP. Range `[0, 1]`. Values of `0` or `1` disable top-p filtering.
+
+  - [`minP`](../../06-api-reference/interfaces/GenerationConfig.md#minp) - Minimum-probability threshold applied after softmax: tokens whose probability is below `minP * max_prob` are excluded from sampling. Range `[0, 1]`. Default `0` disables the filter. Stacks with `topP` when both are set.
+
+  - [`repetitionPenalty`](../../06-api-reference/interfaces/GenerationConfig.md#repetitionpenalty) - Multiplicative penalty applied to logits of tokens that already appeared in the prompt or the generated text. Values greater than `1` discourage repetition; default `1` disables the penalty.
+
+:::info[Built-in models ship with sampling defaults]
+Model presets expose an optional [`generationConfig`](../../06-api-reference/interfaces/LLMProps.md) on the `model` prop. Whenever the upstream model card publishes recommended values (currently Qwen3 and LFM2-VL) the preset carries them and `useLLM` applies them automatically before `isReady` flips — you don't need to call `configure` just to get sensible defaults. Any fields you then pass to `configure` still override on a per-field basis.
+:::
 
 ### Model configuration example
 
@@ -279,7 +287,9 @@ useEffect(() => {
       outputTokenBatchSize: 15,
       batchTimeInterval: 100,
       temperature: 0.7,
-      topp: 0.9,
+      topP: 0.9,
+      minP: 0.05,
+      repetitionPenalty: 1.05,
     },
   });
 }, [configure]);
@@ -488,9 +498,9 @@ Some models support multimodal input — text and images together. To use them,
 ### Loading a VLM
 
 ```tsx
-import { useLLM, LFM2_VL_1_6B_QUANTIZED } from 'react-native-executorch';
+import { useLLM, LFM2_5_VL_1_6B_QUANTIZED } from 'react-native-executorch';
 
-const llm = useLLM({ model: LFM2_VL_1_6B_QUANTIZED });
+const llm = useLLM({ model: LFM2_5_VL_1_6B_QUANTIZED });
 ```
 
 The `capabilities` field is already set on the model constant. You can also construct the model object explicitly:
@@ -511,7 +521,7 @@ Passing `capabilities` unlocks the typed `media` argument on `sendMessage`.
 ### Sending a message with an image
 
 ```tsx
-const llm = useLLM({ model: LFM2_VL_1_6B_QUANTIZED });
+const llm = useLLM({ model: LFM2_5_VL_1_6B_QUANTIZED });
 
 const send = () => {
   llm.sendMessage('What is in this image?', {
@@ -534,7 +544,7 @@ The `imagePath` should be a local file path on the device.
 You can also use `generate` directly by setting `mediaPath` on user messages:
 
 ```tsx
-const llm = useLLM({ model: LFM2_VL_1_6B_QUANTIZED });
+const llm = useLLM({ model: LFM2_5_VL_1_6B_QUANTIZED });
 
 const handleGenerate = async () => {
   const chat: Message[] = [
diff --git a/docs/docs/04-typescript-api/01-natural-language-processing/LLMModule.md b/docs/docs/04-typescript-api/01-natural-language-processing/LLMModule.md
index 3a479e8ae4..4bed548a4e 100644
--- a/docs/docs/04-typescript-api/01-natural-language-processing/LLMModule.md
+++ b/docs/docs/04-typescript-api/01-natural-language-processing/LLMModule.md
@@ -107,17 +107,25 @@ To configure model (i.e. change system prompt, load initial conversation history
 
   - [`temperature`](../../06-api-reference/interfaces/GenerationConfig.md#temperature) - Scales output logits by the inverse of temperature. Controls the randomness / creativity of text generation.
 
-  - [`topp`](../../06-api-reference/interfaces/GenerationConfig.md#topp) - Only samples from the smallest set of tokens whose cumulative probability exceeds topp.
+  - [`topP`](../../06-api-reference/interfaces/GenerationConfig.md#topp) - Only samples from the smallest set of tokens whose cumulative probability exceeds topP. Range `[0, 1]`. Values of `0` or `1` disable top-p filtering.
+
+  - [`minP`](../../06-api-reference/interfaces/GenerationConfig.md#minp) - Minimum-probability threshold applied after softmax: tokens whose probability is below `minP * max_prob` are excluded from sampling. Range `[0, 1]`. Default `0` disables the filter. Stacks with `topP` when both are set.
+
+  - [`repetitionPenalty`](../../06-api-reference/interfaces/GenerationConfig.md#repetitionpenalty) - Multiplicative penalty applied to logits of tokens that already appeared in the prompt or the generated text. Values greater than `1` discourage repetition; default `1` disables the penalty.
+
+:::info[Built-in models ship with sampling defaults]
+Model presets expose an optional `generationConfig` that `LLMModule.fromModelName` applies automatically when available — for Qwen3 and LFM2-VL this means the model-card recommended sampling settings are in effect without any explicit `configure` call. Any fields you pass to `configure` still override on a per-field basis.
+:::
 
 ## Vision-Language Models (VLM)
 
 Some models support multimodal input — text and images together. To use them, pass `capabilities` in the model object when calling [`fromModelName`](../../06-api-reference/classes/LLMModule.md#frommodelname):
 
 ```typescript
-import { LLMModule, LFM2_VL_1_6B_QUANTIZED } from 'react-native-executorch';
+import { LLMModule, LFM2_5_VL_1_6B_QUANTIZED } from 'react-native-executorch';
 
 const llm = await LLMModule.fromModelName(
-  LFM2_VL_1_6B_QUANTIZED,
+  LFM2_5_VL_1_6B_QUANTIZED,
   undefined,
   (token) => console.log(token)
 );
diff --git a/packages/react-native-executorch/common/rnexecutorch/host_objects/ModelHostObject.h b/packages/react-native-executorch/common/rnexecutorch/host_objects/ModelHostObject.h
index 04b0accd34..d311a3de78 100644
--- a/packages/react-native-executorch/common/rnexecutorch/host_objects/ModelHostObject.h
+++ b/packages/react-native-executorch/common/rnexecutorch/host_objects/ModelHostObject.h
@@ -140,6 +140,15 @@ template <typename Model> class ModelHostObject : public JsiHostObject {
                                        synchronousHostFunction<&Model::setTopp>,
                                        "setTopp"));
 
+      addFunctions(JSI_EXPORT_FUNCTION(ModelHostObject<Model>,
+                                       synchronousHostFunction<&Model::setMinP>,
+                                       "setMinP"));
+
+      addFunctions(JSI_EXPORT_FUNCTION(
+          ModelHostObject<Model>,
+          synchronousHostFunction<&Model::setRepetitionPenalty>,
+          "setRepetitionPenalty"));
+
       addFunctions(JSI_EXPORT_FUNCTION(
           ModelHostObject<Model>,
           synchronousHostFunction<&Model::getMaxContextLength>,
diff --git a/packages/react-native-executorch/common/rnexecutorch/models/llm/LLM.cpp b/packages/react-native-executorch/common/rnexecutorch/models/llm/LLM.cpp
index 10aff5cc77..7e0fa4b26e 100644
--- a/packages/react-native-executorch/common/rnexecutorch/models/llm/LLM.cpp
+++ b/packages/react-native-executorch/common/rnexecutorch/models/llm/LLM.cpp
@@ -250,6 +250,30 @@ void LLM::setTopp(float topp) {
   runner_->set_topp(topp);
 }
 
+void LLM::setMinP(float minP) {
+  if (!runner_ || !runner_->is_loaded()) {
+    throw RnExecutorchError(RnExecutorchErrorCode::ModuleNotLoaded,
+                            "Can't configure a model that's not loaded");
+  }
+  if (minP < 0.0f || minP > 1.0f) {
+    throw RnExecutorchError(RnExecutorchErrorCode::InvalidConfig,
+                            "Min-p must be between 0.0 and 1.0");
+  }
+  runner_->set_min_p(minP);
+}
+
+void LLM::setRepetitionPenalty(float repetitionPenalty) {
+  if (!runner_ || !runner_->is_loaded()) {
+    throw RnExecutorchError(RnExecutorchErrorCode::ModuleNotLoaded,
+                            "Can't configure a model that's not loaded");
+  }
+  if (repetitionPenalty < 0.0f) {
+    throw RnExecutorchError(RnExecutorchErrorCode::InvalidConfig,
+                            "Repetition penalty must be non-negative");
+  }
+  runner_->set_repetition_penalty(repetitionPenalty);
+}
+
 int32_t LLM::getMaxContextLength() const {
   if (!runner_ || !runner_->is_loaded()) {
     throw RnExecutorchError(
diff --git a/packages/react-native-executorch/common/rnexecutorch/models/llm/LLM.h b/packages/react-native-executorch/common/rnexecutorch/models/llm/LLM.h
index 5c9bc258d7..222b5bc62f 100644
--- a/packages/react-native-executorch/common/rnexecutorch/models/llm/LLM.h
+++ b/packages/react-native-executorch/common/rnexecutorch/models/llm/LLM.h
@@ -38,6 +38,8 @@ class LLM : public BaseModel {
   void setCountInterval(size_t countInterval);
   void setTemperature(float temperature);
   void setTopp(float topp);
+  void setMinP(float minP);
+  void setRepetitionPenalty(float repetitionPenalty);
   void setTimeInterval(size_t timeInterval);
   int32_t getMaxContextLength() const;
 
diff --git a/packages/react-native-executorch/common/rnexecutorch/tests/CMakeLists.txt b/packages/react-native-executorch/common/rnexecutorch/tests/CMakeLists.txt
index d68ab33509..c3161c777a 100644
--- a/packages/react-native-executorch/common/rnexecutorch/tests/CMakeLists.txt
+++ b/packages/react-native-executorch/common/rnexecutorch/tests/CMakeLists.txt
@@ -149,6 +149,12 @@ add_rn_test(RunnerTests unit/RunnerTest.cpp
         integration/stubs/jsi_stubs.cpp
     LIBS tokenizers_deps
 )
+add_rn_test(SamplerTests unit/SamplerTest.cpp
+    SOURCES
+        ${COMMON_DIR}/runner/sampler.cpp
+        ${COMMON_DIR}/runner/arange_util.cpp
+    LIBS
+)
 add_rn_test(LogTests unit/LogTest.cpp)
 add_rn_test(FileUtilsTest unit/FileUtilsTest.cpp)
 add_rn_test(ImageProcessingTest unit/ImageProcessingTest.cpp
diff --git a/packages/react-native-executorch/common/rnexecutorch/tests/integration/LLMTest.cpp b/packages/react-native-executorch/common/rnexecutorch/tests/integration/LLMTest.cpp
index cad365cc61..ae0a11e777 100644
--- a/packages/react-native-executorch/common/rnexecutorch/tests/integration/LLMTest.cpp
+++ b/packages/react-native-executorch/common/rnexecutorch/tests/integration/LLMTest.cpp
@@ -110,6 +110,31 @@ TEST_F(LLMTest, SetToppInvalidThrows) {
   EXPECT_THROW(model.setTopp(1.1f), RnExecutorchError);
 }
 
+TEST_F(LLMTest, SetMinP) {
+  LLM model(kValidModelPath, kValidTokenizerPath, {}, mockInvoker_);
+  EXPECT_NO_THROW(model.setMinP(0.0f));
+  EXPECT_NO_THROW(model.setMinP(0.15f));
+  EXPECT_NO_THROW(model.setMinP(1.0f));
+}
+
+TEST_F(LLMTest, SetMinPInvalidThrows) {
+  LLM model(kValidModelPath, kValidTokenizerPath, {}, mockInvoker_);
+  EXPECT_THROW(model.setMinP(-0.1f), RnExecutorchError);
+  EXPECT_THROW(model.setMinP(1.1f), RnExecutorchError);
+}
+
+TEST_F(LLMTest, SetRepetitionPenalty) {
+  LLM model(kValidModelPath, kValidTokenizerPath, {}, mockInvoker_);
+  EXPECT_NO_THROW(model.setRepetitionPenalty(1.0f));
+  EXPECT_NO_THROW(model.setRepetitionPenalty(1.05f));
+  EXPECT_NO_THROW(model.setRepetitionPenalty(2.0f));
+}
+
+TEST_F(LLMTest, SetRepetitionPenaltyInvalidThrows) {
+  LLM model(kValidModelPath, kValidTokenizerPath, {}, mockInvoker_);
+  EXPECT_THROW(model.setRepetitionPenalty(-0.1f), RnExecutorchError);
+}
+
 TEST_F(LLMTest, SetCountInterval) {
   LLM model(kValidModelPath, kValidTokenizerPath, {}, mockInvoker_);
   EXPECT_NO_THROW(model.setCountInterval(5));
diff --git a/packages/react-native-executorch/common/rnexecutorch/tests/integration/stubs/StubRunner.h b/packages/react-native-executorch/common/rnexecutorch/tests/integration/stubs/StubRunner.h
index 65f1dde66b..023d6cf080 100644
--- a/packages/react-native-executorch/common/rnexecutorch/tests/integration/stubs/StubRunner.h
+++ b/packages/react-native-executorch/common/rnexecutorch/tests/integration/stubs/StubRunner.h
@@ -18,10 +18,6 @@ class StubRunner : public ::executorch::extension::llm::BaseLLMRunner {
     return ::executorch::runtime::Error::Ok;
   }
   void stop_impl() override {}
-  void set_temperature_impl(float t) override { last_temp_ = t; }
-  void set_topp_impl(float) override {}
-  void set_count_interval_impl(size_t) override {}
-  void set_time_interval_impl(size_t) override {}
 
   int32_t resolve_max(int32_t prompt, int32_t seq_len, int32_t ctx_len,
                       int32_t max_new = -1) const {
@@ -29,5 +25,4 @@ class StubRunner : public ::executorch::extension::llm::BaseLLMRunner {
   }
 
   bool loaded_ = false;
-  float last_temp_ = -1.f;
 };
diff --git a/packages/react-native-executorch/common/rnexecutorch/tests/unit/RunnerTest.cpp b/packages/react-native-executorch/common/rnexecutorch/tests/unit/RunnerTest.cpp
index d7bd344049..77d0d4ae8d 100644
--- a/packages/react-native-executorch/common/rnexecutorch/tests/unit/RunnerTest.cpp
+++ b/packages/react-native-executorch/common/rnexecutorch/tests/unit/RunnerTest.cpp
@@ -62,11 +62,10 @@ TEST(MultimodalInputTest, EmptyStringIsStillText) {
 // BaseLLMRunner via StubRunner
 // ============================================================================
 
-TEST(BaseLLMRunnerTest, SetTemperatureUpdatesConfigAndCallsImpl) {
+TEST(BaseLLMRunnerTest, SetTemperatureUpdatesConfig) {
   StubRunner runner(nullptr, "dummy");
   runner.set_temperature(0.42f);
   EXPECT_FLOAT_EQ(runner.config_.temperature, 0.42f);
-  EXPECT_FLOAT_EQ(runner.last_temp_, 0.42f);
 }
 
 TEST(BaseLLMRunnerTest, SetToppUpdatesConfig) {
@@ -89,3 +88,15 @@ TEST(BaseLLMRunnerTest, GenerateEmptyStringReturnsError) {
   auto err = runner.generate("", {}, {}, {});
   EXPECT_NE(err, ::executorch::runtime::Error::Ok);
 }
+
+TEST(BaseLLMRunnerTest, SetMinPUpdatesConfig) {
+  StubRunner runner(nullptr, "dummy");
+  runner.set_min_p(0.15f);
+  EXPECT_FLOAT_EQ(runner.config_.min_p, 0.15f);
+}
+
+TEST(BaseLLMRunnerTest, SetRepetitionPenaltyUpdatesConfig) {
+  StubRunner runner(nullptr, "dummy");
+  runner.set_repetition_penalty(1.05f);
+  EXPECT_FLOAT_EQ(runner.config_.repetition_penalty, 1.05f);
+}
diff --git a/packages/react-native-executorch/common/rnexecutorch/tests/unit/SamplerTest.cpp b/packages/react-native-executorch/common/rnexecutorch/tests/unit/SamplerTest.cpp
new file mode 100644
index 0000000000..4295f16232
--- /dev/null
+++ b/packages/react-native-executorch/common/rnexecutorch/tests/unit/SamplerTest.cpp
@@ -0,0 +1,85 @@
+/*
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * All rights reserved.
+ *
+ * This source code is licensed under the BSD-style license found in the
+ * LICENSE file in the root directory of this source tree.
+ */
+
+#include <gtest/gtest.h>
+#include <runner/sampler.h>
+#include <vector>
+
+using namespace executorch::extension::llm;
+
+// Helper: run sampler N times, count how often each index is picked.
+template <typename T>
+std::vector<int> sampleMany(Sampler &s, std::vector<T> logits,
+                            const std::vector<uint64_t> &recent, int n) {
+  std::vector<int> counts(logits.size(), 0);
+  for (int i = 0; i < n; ++i) {
+    std::vector<T> copy = logits;
+    counts[s.sample(copy.data(), recent)]++;
+  }
+  return counts;
+}
+
+// 1. Repetition penalty on positive logit: token 0 should be sampled less.
+TEST(SamplerTest, RepetitionPenaltyReducesPositiveLogit) {
+  Sampler s(2, 1.0f, 1.0f, 0, 0.0f, 1.3f);
+  std::vector<float> logits = {1.0f, 1.0f};
+  std::vector<uint64_t> recent = {0};
+  auto counts = sampleMany(s, logits, recent, 2000);
+  EXPECT_LT(counts[0], 1200);
+}
+
+// 2. Repetition penalty on negative logit: penalised token should appear even
+// less.
+TEST(SamplerTest, RepetitionPenaltyMultipliesNegativeLogit) {
+  Sampler s(2, 1.0f, 1.0f, 0, 0.0f, 1.5f);
+  std::vector<float> logits = {0.0f, -1.0f};
+  std::vector<uint64_t> recent = {1};
+  auto counts = sampleMany(s, logits, recent, 2000);
+  EXPECT_LT(counts[1], 200);
+}
+
+// 3. No recent tokens — penalty has no effect.
+TEST(SamplerTest, RepetitionPenaltyNoRecentTokensHasNoEffect) {
+  Sampler baseline(2, 1.0f, 1.0f, 0, 0.0f, 1.0f);
+  Sampler penalised(2, 1.0f, 1.0f, 0, 0.0f, 2.0f);
+  std::vector<float> logits_b = {1.0f, 1.0f};
+  std::vector<float> logits_p = {1.0f, 1.0f};
+  std::vector<uint64_t> recent = {};
+  auto cb = sampleMany(baseline, logits_b, recent, 2000);
+  auto cp = sampleMany(penalised, logits_p, recent, 2000);
+  EXPECT_NEAR(cb[0], cp[0], 300);
+}
+
+// 4. Min-p truncation: token with very low probability is excluded.
+TEST(SamplerTest, MinPFiltersTailTokens) {
+  Sampler s(3, 1.0f, 1.0f, 0, 0.1f, 1.0f);
+  std::vector<float> logits = {5.0f, -5.0f, -5.0f};
+  std::vector<uint64_t> recent = {};
+  auto counts = sampleMany(s, logits, recent, 1000);
+  EXPECT_EQ(counts[1], 0);
+  EXPECT_EQ(counts[2], 0);
+  EXPECT_EQ(counts[0], 1000);
+}
+
+// 5. Min-p = 0 disables filtering.
+TEST(SamplerTest, MinPZeroDisablesFiltering) {
+  Sampler s(3, 0.0f, 1.0f, 0, 0.0f, 1.0f);
+  std::vector<float> logits = {1.0f, -1000.0f, -1000.0f};
+  std::vector<uint64_t> recent = {};
+  EXPECT_EQ(s.sample(logits.data(), recent), 0);
+}
+
+// 6. Min-p + top-p stacked.
+TEST(SamplerTest, MinPAndToppStack) {
+  Sampler s(4, 1.0f, 0.5f, 0, 0.2f, 1.0f);
+  std::vector<float> logits = {5.0f, 2.0f, -2.0f, -5.0f};
+  std::vector<uint64_t> recent = {};
+  auto counts = sampleMany(s, logits, recent, 2000);
+  EXPECT_EQ(counts[2], 0);
+  EXPECT_EQ(counts[3], 0);
+}
diff --git a/packages/react-native-executorch/common/runner/base_llm_runner.cpp b/packages/react-native-executorch/common/runner/base_llm_runner.cpp
index d38a41b76b..a021040807 100644
--- a/packages/react-native-executorch/common/runner/base_llm_runner.cpp
+++ b/packages/react-native-executorch/common/runner/base_llm_runner.cpp
@@ -139,20 +139,22 @@ int32_t BaseLLMRunner::get_max_context_length() const {
 
 void BaseLLMRunner::set_temperature(float temperature) noexcept {
   config_.temperature = temperature;
-  set_temperature_impl(temperature);
 }
 
-void BaseLLMRunner::set_topp(float topp) noexcept {
-  config_.topp = topp;
-  set_topp_impl(topp);
+void BaseLLMRunner::set_topp(float topp) noexcept { config_.topp = topp; }
+
+void BaseLLMRunner::set_min_p(float min_p) noexcept { config_.min_p = min_p; }
+
+void BaseLLMRunner::set_repetition_penalty(float repetition_penalty) noexcept {
+  config_.repetition_penalty = repetition_penalty;
 }
 
 void BaseLLMRunner::set_count_interval(size_t count_interval) {
-  set_count_interval_impl(count_interval);
+  config_.output_token_batch_size = count_interval;
 }
 
 void BaseLLMRunner::set_time_interval(size_t time_interval) {
-  set_time_interval_impl(time_interval);
+  config_.batch_time_interval_ms = time_interval;
 }
 
 int32_t BaseLLMRunner::resolve_max_new_tokens(int32_t num_prompt_tokens,
diff --git a/packages/react-native-executorch/common/runner/base_llm_runner.h b/packages/react-native-executorch/common/runner/base_llm_runner.h
index 3924aa3d7a..9710f5ae70 100644
--- a/packages/react-native-executorch/common/runner/base_llm_runner.h
+++ b/packages/react-native-executorch/common/runner/base_llm_runner.h
@@ -53,6 +53,8 @@ class BaseLLMRunner {
 
   void set_temperature(float temperature) noexcept;
   void set_topp(float topp) noexcept;
+  void set_min_p(float min_p) noexcept;
+  void set_repetition_penalty(float repetition_penalty) noexcept;
   void set_count_interval(size_t count_interval);
   void set_time_interval(size_t time_interval);
 
@@ -65,10 +67,12 @@ class BaseLLMRunner {
 protected:
   virtual ::executorch::runtime::Error load_subcomponents() = 0;
   virtual void stop_impl() = 0;
-  virtual void set_temperature_impl(float temperature) = 0;
-  virtual void set_topp_impl(float topp) = 0;
-  virtual void set_count_interval_impl(size_t count_interval) = 0;
-  virtual void set_time_interval_impl(size_t time_interval) = 0;
+  // Sampling values and token-batching intervals live entirely in `config_`.
+  // The TextDecoderRunner / TextTokenGenerator shared by both TextRunner and
+  // MultimodalRunner are constructed with a const reference to `config_`
+  // and read those fields on every iteration, so writes via the public
+  // set_* methods on BaseLLMRunner take effect immediately with no virtual
+  // dispatch needed.
 
   int32_t resolve_max_new_tokens(int32_t num_prompt_tokens, int32_t max_seq_len,
                                  int32_t max_context_len,
diff --git a/packages/react-native-executorch/common/runner/encoders/vision_encoder.cpp b/packages/react-native-executorch/common/runner/encoders/vision_encoder.cpp
index 4f664a173e..de3e196c1f 100644
--- a/packages/react-native-executorch/common/runner/encoders/vision_encoder.cpp
+++ b/packages/react-native-executorch/common/runner/encoders/vision_encoder.cpp
@@ -77,15 +77,23 @@ Result<VisionEncoder::ImageShape> VisionEncoder::getInputShape() const {
 std::vector<float>
 VisionEncoder::preprocessImage(const std::string &path,
                                const ImageShape &targetShape) const {
-  cv::Mat mat = rnexecutorch::image_processing::readImage(path);
-  cv::resize(mat, mat, cv::Size(targetShape.width, targetShape.height));
-  cv::cvtColor(mat, mat, cv::COLOR_BGR2RGB);
+  // The bundled vision-encoder PTEs (e.g. LFM2.5-VL) bake rescale + normalize
+  // into the exported graph, so we hand raw 0-255 float pixel values to the
+  // module. Adding rescale / normalize here would double-apply the transform
+  // and destroy the input distribution. We reuse `resizePadded` for the
+  // aspect-ratio-preserving letterbox (it picks the pad colour from the
+  // source image corners, which blends better than a flat gray), then
+  // convert BGR->RGB and repack the raw pixels into CHW float.
+  cv::Mat src = rnexecutorch::image_processing::readImage(path);
+  cv::Mat canvas = rnexecutorch::image_processing::resizePadded(
+      src, cv::Size(targetShape.width, targetShape.height));
+  cv::cvtColor(canvas, canvas, cv::COLOR_BGR2RGB);
 
   const int32_t pixelCount = targetShape.height * targetShape.width;
   std::vector<float> chw(targetShape.channels * pixelCount);
   for (int32_t i = 0; i < pixelCount; ++i) {
     cv::Vec3b px =
-        mat.at<cv::Vec3b>(i / targetShape.width, i % targetShape.width);
+        canvas.at<cv::Vec3b>(i / targetShape.width, i % targetShape.width);
     for (int32_t c = 0; c < targetShape.channels; ++c) {
       chw[c * pixelCount + i] = static_cast<float>(px[c]);
     }
diff --git a/packages/react-native-executorch/common/runner/irunner.h b/packages/react-native-executorch/common/runner/irunner.h
index 8dedc6687e..54b14c354f 100644
--- a/packages/react-native-executorch/common/runner/irunner.h
+++ b/packages/react-native-executorch/common/runner/irunner.h
@@ -58,6 +58,21 @@ struct GenerationConfig {
   // = more deterministic, higher = more diverse generations.
   float topp = -1.F;
 
+  // Minimum probability threshold: tokens with prob < min_p * max_prob are
+  // excluded. 0.0 disables min_p filtering.
+  float min_p = 0.0f;
+
+  // Multiplicative penalty applied to logits of recently generated tokens.
+  // Values > 1.0 discourage repetition. 1.0 disables the penalty.
+  float repetition_penalty = 1.0f;
+
+  // Token-batching parameters for the streaming token callback. The
+  // generator flushes a batch when either `output_token_batch_size` tokens
+  // have accumulated or `batch_time_interval_ms` milliseconds have elapsed
+  // since the last flush, whichever comes first.
+  size_t output_token_batch_size = 10;
+  size_t batch_time_interval_ms = 120;
+
   // Enable dynamic input shapes (if implemented) or not
   // Impacts the prefill phase and causes TextPrefiller to pass all the tokens
   // at once if set to true.
diff --git a/packages/react-native-executorch/common/runner/multimodal_decoder_runner.h b/packages/react-native-executorch/common/runner/multimodal_decoder_runner.h
index e6fa4e7eae..071b193539 100644
--- a/packages/react-native-executorch/common/runner/multimodal_decoder_runner.h
+++ b/packages/react-native-executorch/common/runner/multimodal_decoder_runner.h
@@ -16,8 +16,9 @@
 namespace executorch::extension::llm {
 class MultimodalDecoderRunner : public TextDecoderRunner {
 public:
-  explicit MultimodalDecoderRunner(Module &module, IOManager *io_manager)
-      : TextDecoderRunner(module, io_manager) {}
+  explicit MultimodalDecoderRunner(Module &module, IOManager *io_manager,
+                                   const GenerationConfig &config)
+      : TextDecoderRunner(module, io_manager, config) {}
 
   inline ::executorch::runtime::Result<::executorch::aten::Tensor>
   step(TensorPtr &tokens, int64_t start_pos) override {
diff --git a/packages/react-native-executorch/common/runner/multimodal_runner.cpp b/packages/react-native-executorch/common/runner/multimodal_runner.cpp
index c81d15760e..767fef9f38 100644
--- a/packages/react-native-executorch/common/runner/multimodal_runner.cpp
+++ b/packages/react-native-executorch/common/runner/multimodal_runner.cpp
@@ -47,8 +47,8 @@ Error MultimodalRunner::load_subcomponents() {
 
   Stats *stats_ptr = &stats_;
 
-  mm_decoder_runner_ =
-      std::make_unique<MultimodalDecoderRunner>(*module_, io_manager_.get());
+  mm_decoder_runner_ = std::make_unique<MultimodalDecoderRunner>(
+      *module_, io_manager_.get(), config_);
   IEncoder *image_encoder = nullptr;
   auto enc_it = encoders_.find(MultimodalType::Image);
   if (enc_it != encoders_.end()) {
@@ -58,7 +58,7 @@ Error MultimodalRunner::load_subcomponents() {
       *module_, *mm_decoder_runner_, *tokenizer_, image_encoder);
   mm_token_generator_ = std::make_unique<TextTokenGenerator>(
       tokenizer_.get(), mm_decoder_runner_.get(), /*use_kv_cache=*/true,
-      std::move(eos_ids_), stats_ptr);
+      std::move(eos_ids_), stats_ptr, config_);
 
   ET_CHECK_OK_OR_RETURN_ERROR(mm_prefiller_->load());
   ET_CHECK_OK_OR_RETURN_ERROR(mm_token_generator_->load());
@@ -106,7 +106,7 @@ Error MultimodalRunner::generate_internal(
   auto generate_result = mm_token_generator_->generate(
       seed_tokens, pos_,
       static_cast<uint64_t>(std::max(0, resolved_max_new - 1)),
-      config_.temperature, config_.topp, wrapped_callback);
+      wrapped_callback);
 
   if (!generate_result.ok())
     return generate_result.error();
@@ -125,16 +125,4 @@ void MultimodalRunner::stop_impl() {
   }
 }
 
-void MultimodalRunner::set_count_interval_impl(size_t count_interval) {
-  if (mm_token_generator_) {
-    mm_token_generator_->set_count_interval(count_interval);
-  }
-}
-
-void MultimodalRunner::set_time_interval_impl(size_t time_interval) {
-  if (mm_token_generator_) {
-    mm_token_generator_->set_time_interval(time_interval);
-  }
-}
-
 } // namespace executorch::extension::llm
diff --git a/packages/react-native-executorch/common/runner/multimodal_runner.h b/packages/react-native-executorch/common/runner/multimodal_runner.h
index 3c31c0165f..d24e0b40c2 100644
--- a/packages/react-native-executorch/common/runner/multimodal_runner.h
+++ b/packages/react-native-executorch/common/runner/multimodal_runner.h
@@ -30,10 +30,6 @@ class MultimodalRunner : public BaseLLMRunner {
 protected:
   ::executorch::runtime::Error load_subcomponents() override;
   void stop_impl() override;
-  void set_temperature_impl(float) override {}
-  void set_topp_impl(float) override {}
-  void set_count_interval_impl(size_t count_interval) override;
-  void set_time_interval_impl(size_t time_interval) override;
 
 private:
   std::map<MultimodalType, std::unique_ptr<IEncoder>> encoders_;
diff --git a/packages/react-native-executorch/common/runner/sampler.cpp b/packages/react-native-executorch/common/runner/sampler.cpp
index a5fadef93c..26c75d4dd5 100644
--- a/packages/react-native-executorch/common/runner/sampler.cpp
+++ b/packages/react-native-executorch/common/runner/sampler.cpp
@@ -35,6 +35,7 @@
 #include "sampler.h"
 #include <algorithm>
 #include <ctime>
+#include <vector>
 
 namespace executorch {
 namespace extension {
@@ -119,16 +120,16 @@ int32_t Sampler::sample_topp(T *probabilities, float coin) {
   return probindex[last_idx].index; // in case of rounding errors
 }
 
-Sampler::Sampler(int vocab_size, float temperature, float topp,
-                 unsigned long long rng_seed)
+Sampler::Sampler(int32_t vocab_size, float temperature, float topp,
+                 unsigned long long rng_seed, float min_p,
+                 float repetition_penalty)
     : vocab_size_(vocab_size),
       inv_temperature_((temperature != 0.0f) ? (1.0f / temperature) : 0.0f),
-      topp_(topp), rng_state_(rng_seed) {}
+      topp_(topp), min_p_(min_p), repetition_penalty_(repetition_penalty),
+      rng_state_(rng_seed) {}
 
 Sampler::Sampler(int vocab_size, float temperature, float topp)
-    : vocab_size_(vocab_size),
-      inv_temperature_((temperature != 0.0f) ? (1.0f / temperature) : 0.0f),
-      topp_(topp), rng_state_(std::time(nullptr)) {}
+    : Sampler(vocab_size, temperature, topp, std::time(nullptr), 0.0f, 1.0f) {}
 
 template <typename T> static void softmax(T *x, int size) {
   // find max value (for numerical stability)
@@ -162,22 +163,25 @@ static float random_f32(unsigned long long *state) { // random float32 in [0,1)
   return (random_u32(state) >> 8) / 16777216.0f;
 }
 
-template <typename T> int32_t Sampler::sample(T *logits) {
+template <typename T>
+int32_t Sampler::sample(T *logits, const std::vector<uint64_t> &recent_tokens) {
   // sample the token given the logits and some hyperparameters
   int next;
   if (inv_temperature_ == 0.0f) {
     // greedy argmax sampling: take the token with the highest probability
     next = sample_argmax(logits);
   } else {
-    // apply the temperature to the logits
-    for (int q = 0; q < vocab_size_; q++) {
-      logits[q] *= inv_temperature_;
-    }
-    // apply softmax to the logits to get the probabilities for next token
+    // 1. apply repetition penalty to raw logits (pre-softmax)
+    apply_repetition_penalty(logits, vocab_size_, recent_tokens);
+    // 2. apply the temperature to the logits
+    apply_temperature(logits, vocab_size_);
+    // 3. apply softmax to the logits to get the probabilities for next token
     softmax(logits, vocab_size_);
+    // 4. apply min_p truncation
+    apply_min_p(logits, vocab_size_);
     // flip a (float) coin (this is our source of entropy for sampling)
     float coin = random_f32(&rng_state_);
-    // we sample from this distribution to get the next token
+    // 5. we sample from this distribution to get the next token
     if (topp_ <= 0 || topp_ >= 1) {
       // simply sample from the predicted probability distribution
       next = sample_mult(logits, coin);
@@ -189,6 +193,10 @@ template <typename T> int32_t Sampler::sample(T *logits) {
   return next;
 }
 
+template <typename T> int32_t Sampler::sample(T *logits) {
+  return sample(logits, {});
+}
+
 template int32_t Sampler::sample<float>(float *logits);
 template int32_t Sampler::sample<uint16_t>(uint16_t *logits);
 template int32_t
@@ -196,6 +204,17 @@ Sampler::sample<executorch::aten::Half>(executorch::aten::Half *logits);
 template int32_t
 Sampler::sample<executorch::aten::BFloat16>(executorch::aten::BFloat16 *logits);
 
+template int32_t Sampler::sample<float>(float *logits,
+                                        const std::vector<uint64_t> &);
+template int32_t Sampler::sample<uint16_t>(uint16_t *logits,
+                                           const std::vector<uint64_t> &);
+template int32_t
+Sampler::sample<executorch::aten::Half>(executorch::aten::Half *logits,
+                                        const std::vector<uint64_t> &);
+template int32_t
+Sampler::sample<executorch::aten::BFloat16>(executorch::aten::BFloat16 *logits,
+                                            const std::vector<uint64_t> &);
+
 } // namespace llm
 } // namespace extension
 } // namespace executorch
diff --git a/packages/react-native-executorch/common/runner/sampler.h b/packages/react-native-executorch/common/runner/sampler.h
index a46a5ed12c..16811297ef 100644
--- a/packages/react-native-executorch/common/runner/sampler.h
+++ b/packages/react-native-executorch/common/runner/sampler.h
@@ -8,12 +8,15 @@
 
 #pragma once
 
+#include <algorithm>
 #include <cctype>
 #include <cmath>
 #include <cstdio>
 #include <cstdlib>
 #include <cstring>
 #include <memory>
+#include <utility>
+#include <vector>
 #ifdef USE_ATEN_LIB
 #include <torch/torch.h>
 #endif
@@ -36,22 +39,77 @@ template <typename T> struct ProbIndex {
 class Sampler {
 public:
   Sampler(int32_t vocab_size, float temperature, float topp,
-          unsigned long long rng_seed);
+          unsigned long long rng_seed, float min_p = 0.0f,
+          float repetition_penalty = 1.0f);
 
   Sampler(int32_t vocab_size, float temperature, float topp);
 
   template <typename T> int32_t sample(T *logits);
 
+  template <typename T>
+  int32_t sample(T *logits, const std::vector<uint64_t> &recent_tokens);
+
 private:
   template <typename T> int32_t sample_topp(T *probabilities, float coin);
   template <typename T> int32_t sample_mult(T *probabilities, float coin);
   template <typename T> int32_t sample_argmax(T *probabilities);
 
+  template <typename T>
+  inline void apply_temperature(T *logits, int32_t vocab_size) {
+    for (std::size_t i = 0; std::cmp_less(i, vocab_size); ++i) {
+      logits[i] =
+          static_cast<T>(static_cast<float>(logits[i]) * inv_temperature_);
+    }
+  }
+
+  template <typename T>
+  inline void
+  apply_repetition_penalty(T *logits, int32_t vocab_size,
+                           const std::vector<uint64_t> &recent_tokens) {
+    if (repetition_penalty_ == 1.0f || recent_tokens.empty())
+      return;
+    for (uint64_t id : recent_tokens) {
+      if (!std::cmp_less(id, vocab_size)) {
+        continue;
+      }
+      T &val = logits[id];
+      if (val > T(0)) {
+        val = static_cast<T>(static_cast<float>(val) / repetition_penalty_);
+      } else {
+        val = static_cast<T>(static_cast<float>(val) * repetition_penalty_);
+      }
+    }
+  }
+
+  template <typename T>
+  inline void apply_min_p(T *probabilities, int32_t vocab_size) {
+    if (min_p_ <= 0.0f) {
+      return;
+    }
+    T max_prob = *std::max_element(probabilities, probabilities + vocab_size);
+    T threshold = static_cast<T>(min_p_ * static_cast<float>(max_prob));
+    T sum = T(0);
+    for (std::size_t i = 0; std::cmp_less(i, vocab_size); ++i) {
+      if (probabilities[i] < threshold) {
+        probabilities[i] = T(0);
+      } else {
+        sum += probabilities[i];
+      }
+    }
+    if (sum > T(0)) {
+      for (std::size_t i = 0; std::cmp_less(i, vocab_size); ++i) {
+        probabilities[i] /= sum;
+      }
+    }
+  }
+
 private:
   int32_t vocab_size_;
   // reciprocal of temperature, or 0 if temperature == 0.
   float inv_temperature_;
   float topp_;
+  float min_p_;
+  float repetition_penalty_;
   unsigned long long rng_state_;
 };
 
diff --git a/packages/react-native-executorch/common/runner/text_decoder_runner.cpp b/packages/react-native-executorch/common/runner/text_decoder_runner.cpp
index fdd9e4489c..e67d3e41fb 100644
--- a/packages/react-native-executorch/common/runner/text_decoder_runner.cpp
+++ b/packages/react-native-executorch/common/runner/text_decoder_runner.cpp
@@ -10,6 +10,7 @@
 
 #include "text_decoder_runner.h"
 #include "arange_util.h"
+#include "irunner.h"
 #include "stats.h"
 
 #include <ctime>
@@ -22,9 +23,8 @@ namespace llm {
 // and a ~5% improvement on Galaxy S22 by switching to
 // FileDataLoader instead of MmapDataLoader + UseMlockIgnoreErrors.
 TextDecoderRunner::TextDecoderRunner(Module &module, IOManager *io_manager,
-                                     float temperature, float topp)
-    : module_(&module), io_manager_(io_manager), temperature_(temperature),
-      topp_(topp) {}
+                                     const GenerationConfig &config)
+    : module_(&module), io_manager_(io_manager), config_(config) {}
 
 // This function is functional, meaning it shouldn't modify any state of the
 // input. It should be safe to call multiple times with the same inputs. The
@@ -82,6 +82,34 @@ TextDecoderRunner::step(TensorPtr &tokens, int64_t start_pos) {
   }
 }
 
+int32_t TextDecoderRunner::logits_to_token(
+    const executorch::aten::Tensor &logits_tensor,
+    const std::vector<uint64_t> &recent_tokens) {
+  int32_t result = 0;
+
+  struct {
+    [[noreturn]] void fail(torch::executor::Error) {
+      ET_CHECK_MSG(false, "Unsupported dtype in logits_to_token");
+    }
+  } ctx;
+
+  ET_SWITCH_FOUR_TYPES(
+      Float, Half, BFloat16, UInt16, logits_tensor.scalar_type(), ctx,
+      "logits_to_token", CTYPE, [&]() {
+        auto *logits = logits_tensor.mutable_data_ptr<CTYPE>();
+        ssize_t vocab_size = logits_tensor.size(logits_tensor.dim() - 1);
+        if (logits_tensor.dim() == 3) {
+          auto num_tokens = logits_tensor.size(1);
+          logits += (num_tokens - 1) * vocab_size;
+        }
+        Sampler sampler(vocab_size, config_.temperature, config_.topp,
+                        static_cast<unsigned long long>(std::time(nullptr)),
+                        config_.min_p, config_.repetition_penalty);
+        result = sampler.sample(logits, recent_tokens);
+      });
+  return result;
+}
+
 } // namespace llm
 } // namespace extension
 } // namespace executorch
diff --git a/packages/react-native-executorch/common/runner/text_decoder_runner.h b/packages/react-native-executorch/common/runner/text_decoder_runner.h
index bd318da526..bffc254bd6 100644
--- a/packages/react-native-executorch/common/runner/text_decoder_runner.h
+++ b/packages/react-native-executorch/common/runner/text_decoder_runner.h
@@ -17,10 +17,15 @@ namespace executorch {
 namespace extension {
 namespace llm {
 
+// Forward declaration to avoid the include chain
+// irunner.h -> stats.h -> util.h -> text_prefiller.h -> text_decoder_runner.h
+// which would re-enter this header before TextDecoderRunner is defined.
+struct GenerationConfig;
+
 class TextDecoderRunner {
 public:
   explicit TextDecoderRunner(Module &module, IOManager *io_manager,
-                             float temperature = 0.8F, float topp = 0.9F);
+                             const GenerationConfig &config);
 
   virtual ~TextDecoderRunner() = default;
 
@@ -50,61 +55,23 @@ class TextDecoderRunner {
     return module_->is_method_loaded("forward");
   }
 
-  virtual void set_temperature(float temperature) noexcept {
-    temperature_ = temperature;
-  }
-
-  virtual void set_topp(float topp) noexcept { topp_ = topp; }
-
   inline void stop() { should_stop_ = true; }
 
   /**
-   * Sample the next token from the logits tensor.
-   * @param logits_tensor The logits tensor.
-   * @param temperature The temperature parameter used to control randomness in
-   * sampling.
-   * @return The next token.
+   * Sample the next token from the logits tensor using the sampling
+   * parameters in `config_`. `recent_tokens` is the prompt-plus-generated
+   * window used by the repetition penalty. Defined out-of-line in the cpp
+   * so this header doesn't need a complete `GenerationConfig` type.
    */
-  inline int32_t logits_to_token(const executorch::aten::Tensor &logits_tensor,
-                                 float temperature = -1.F, float topp = -1.F) {
-    int32_t result = 0;
-
-    temperature = temperature < 0.F ? temperature_ : temperature;
-    topp = topp < 0.F ? topp_ : topp;
-
-    // Create a minimal context for error handling in ET_SWITCH
-    struct {
-      [[noreturn]] void fail(torch::executor::Error /* error */) {
-        ET_CHECK_MSG(false, "Unsupported dtype in logits_to_token");
-      }
-    } ctx;
-
-    ET_SWITCH_FOUR_TYPES(
-        Float, Half, BFloat16, UInt16, logits_tensor.scalar_type(), ctx,
-        "logits_to_token", CTYPE, [&]() {
-          // If the logit_tensor rank is 3, the shape is [batch, seq_length,
-          // vocab_size], get the last logits, sample and return. Else the model
-          // outputs the last logit, directly sample and return.
-          auto *logits = logits_tensor.mutable_data_ptr<CTYPE>();
-          ssize_t vocab_size = logits_tensor.size(logits_tensor.dim() - 1);
-          if (logits_tensor.dim() == 3) {
-            auto num_tokens = logits_tensor.size(1);
-            logits += (num_tokens - 1) * vocab_size;
-          }
-          // @lint-ignore CLANGTIDY facebook-hte-Deprecated
-          Sampler sampler(vocab_size, temperature, topp);
-          result = sampler.sample(logits);
-        });
-    return result;
-  }
+  int32_t logits_to_token(const executorch::aten::Tensor &logits_tensor,
+                          const std::vector<uint64_t> &recent_tokens = {});
 
 protected:
   // Non-owning. The runner (BaseLLMRunner) owns the Module and outlives this.
   Module *module_;
   IOManager *io_manager_;
   bool should_stop_{false};
-  float temperature_;
-  float topp_;
+  const GenerationConfig &config_;
 };
 
 } // namespace llm
diff --git a/packages/react-native-executorch/common/runner/text_runner.cpp b/packages/react-native-executorch/common/runner/text_runner.cpp
index 6a0e202084..5a75e00b4a 100644
--- a/packages/react-native-executorch/common/runner/text_runner.cpp
+++ b/packages/react-native-executorch/common/runner/text_runner.cpp
@@ -27,13 +27,13 @@ Error TextRunner::load_subcomponents() {
   Stats *stats_ptr = &stats_;
 
   text_decoder_runner_ = std::make_unique<TextDecoderRunner>(
-      *module_, io_manager_.get(), config_.temperature, config_.topp);
+      *module_, io_manager_.get(), config_);
   text_prefiller_ = std::make_unique<TextPrefiller>(
       text_decoder_runner_.get(), config_.enable_kv_cache,
       config_.enable_dynamic_shape, config_.max_seq_len);
   text_token_generator_ = std::make_unique<TextTokenGenerator>(
       tokenizer_.get(), text_decoder_runner_.get(), config_.enable_kv_cache,
-      std::move(eos_ids_), stats_ptr);
+      std::move(eos_ids_), stats_ptr, config_);
 
   return Error::Ok;
 }
@@ -112,8 +112,7 @@ Error TextRunner::generate_internal(
 
   prompt_tokens.push_back(cur_token);
   int64_t num_generated = ET_UNWRAP(text_token_generator_->generate(
-      prompt_tokens, pos_, max_new_tokens - 1, config_.temperature,
-      config_.topp, wrapped_callback));
+      prompt_tokens, pos_, max_new_tokens - 1, wrapped_callback));
 
   pos_ += num_generated;
   stats_.inference_end_ms = time_in_ms();
@@ -128,26 +127,4 @@ void TextRunner::stop_impl() {
     text_token_generator_->stop();
 }
 
-void TextRunner::set_temperature_impl(float temperature) {
-  if (text_decoder_runner_)
-    text_decoder_runner_->set_temperature(temperature);
-}
-
-void TextRunner::set_topp_impl(float topp) {
-  if (text_decoder_runner_)
-    text_decoder_runner_->set_topp(topp);
-}
-
-void TextRunner::set_count_interval_impl(size_t count_interval) {
-  if (text_token_generator_) {
-    text_token_generator_->set_count_interval(count_interval);
-  }
-}
-
-void TextRunner::set_time_interval_impl(size_t time_interval) {
-  if (text_token_generator_) {
-    text_token_generator_->set_time_interval(time_interval);
-  }
-}
-
 } // namespace executorch::extension::llm
diff --git a/packages/react-native-executorch/common/runner/text_runner.h b/packages/react-native-executorch/common/runner/text_runner.h
index 4fce0e8150..096d3498d7 100644
--- a/packages/react-native-executorch/common/runner/text_runner.h
+++ b/packages/react-native-executorch/common/runner/text_runner.h
@@ -24,10 +24,6 @@ class TextRunner : public BaseLLMRunner {
 protected:
   ::executorch::runtime::Error load_subcomponents() override;
   void stop_impl() override;
-  void set_temperature_impl(float temperature) override;
-  void set_topp_impl(float topp) override;
-  void set_count_interval_impl(size_t count_interval) override;
-  void set_time_interval_impl(size_t time_interval) override;
 
 private:
   std::unique_ptr<TextDecoderRunner> text_decoder_runner_;
diff --git a/packages/react-native-executorch/common/runner/text_token_generator.h b/packages/react-native-executorch/common/runner/text_token_generator.h
index 024cce456e..7ecf6177a9 100644
--- a/packages/react-native-executorch/common/runner/text_token_generator.h
+++ b/packages/react-native-executorch/common/runner/text_token_generator.h
@@ -9,6 +9,7 @@
 // Generate tokens in a loop.
 #pragma once
 
+#include "irunner.h"
 #include "stats.h"
 #include "text_decoder_runner.h"
 #include "util.h"
@@ -22,13 +23,18 @@ namespace llm {
 
 class TextTokenGenerator {
 public:
+  // Holds a const reference to the owning runner's GenerationConfig and
+  // reads `output_token_batch_size` / `batch_time_interval_ms` from it on
+  // every iteration of the generation loop, so external updates take effect
+  // mid-stream without any sync setter on this class.
   TextTokenGenerator(tokenizers::HFTokenizer *tokenizer,
                      TextDecoderRunner *text_decoder_runner, bool use_kv_cache,
                      std::unique_ptr<std::unordered_set<uint64_t>> &&eos_ids,
-                     Stats *stats)
+                     Stats *stats, const GenerationConfig &config)
       : tokenizer_(tokenizer), text_decoder_runner_(text_decoder_runner),
         eos_ids_(std::move(eos_ids)), use_kv_cache_(use_kv_cache),
-        timestamp_(std::chrono::high_resolution_clock::now()), stats_(stats) {}
+        timestamp_(std::chrono::high_resolution_clock::now()), stats_(stats),
+        config_(config) {}
 
   virtual ~TextTokenGenerator() = default;
 
@@ -48,7 +54,6 @@ class TextTokenGenerator {
    */
   inline ::executorch::runtime::Result<int64_t> generate(
       std::vector<uint64_t> tokens, int64_t start_pos, uint64_t max_new_tokens,
-      float temperature, float topp,
       const std::function<void(const std::string &)> &token_callback = {}) {
     ET_CHECK_MSG(!tokens.empty(),
                  "Token generation loop shouldn't take empty tokens");
@@ -80,6 +85,7 @@ class TextTokenGenerator {
     auto tokens_managed = from_blob(token_data.data(), token_shape,
                                     executorch::aten::ScalarType::Long);
 
+    std::vector<uint64_t> generated_tokens(tokens.begin(), tokens.end());
     should_stop_ = false;
     timestamp_ = std::chrono::high_resolution_clock::now();
 
@@ -94,11 +100,12 @@ class TextTokenGenerator {
       prev_token = cur_token;
 
       stats_->on_sampling_begin();
-      cur_token = text_decoder_runner_->logits_to_token(logits_tensor,
-                                                        temperature, topp);
+      cur_token =
+          text_decoder_runner_->logits_to_token(logits_tensor, generated_tokens);
       stats_->on_sampling_end();
 
       pos++;
+      generated_tokens.push_back(cur_token);
 
       if (use_kv_cache_) {
         // update the token tensor. token_data will not be empty.
@@ -128,9 +135,10 @@ class TextTokenGenerator {
       std::string cache_decoded = decodeResult.get();
       const auto timeIntervalElapsed =
           std::chrono::duration_cast<std::chrono::milliseconds>(
-              std::chrono::high_resolution_clock::now() - timestamp_) >
-          time_interval_;
-      const auto countIntervalElapsed = token_cache.size() > count_interval_;
+              std::chrono::high_resolution_clock::now() - timestamp_)
+              .count() > static_cast<int64_t>(config_.batch_time_interval_ms);
+      const auto countIntervalElapsed =
+          token_cache.size() > config_.output_token_batch_size;
       const auto eos_reached = eos_ids_->contains(cur_token);
 
       if (!cache_decoded.ends_with("�") &&
@@ -177,14 +185,6 @@ class TextTokenGenerator {
     return text_decoder_runner_->is_method_loaded();
   }
 
-  void set_count_interval(size_t count_interval) {
-    count_interval_ = count_interval;
-  }
-
-  void set_time_interval(size_t time_interval) {
-    time_interval_ = std::chrono::milliseconds(time_interval);
-  }
-
 private:
   /**
    * Note: TextTokenGenerator does not own the tokenizer_ and
@@ -199,12 +199,14 @@ class TextTokenGenerator {
 
   // state machine
   bool should_stop_ = false;
-  size_t count_interval_{10};
-  std::chrono::milliseconds time_interval_{120};
   std::chrono::high_resolution_clock::time_point timestamp_;
 
   // stats
   Stats *stats_;
+
+  // Reference to the owning runner's GenerationConfig. Token-batching
+  // intervals are read fresh on every iteration.
+  const GenerationConfig &config_;
 };
 
 } // namespace llm
diff --git a/packages/react-native-executorch/src/constants/modelUrls.ts b/packages/react-native-executorch/src/constants/modelUrls.ts
index 3bf07a1758..9cb2689fe2 100644
--- a/packages/react-native-executorch/src/constants/modelUrls.ts
+++ b/packages/react-native-executorch/src/constants/modelUrls.ts
@@ -83,6 +83,14 @@ const QWEN3_4B_QUANTIZED_MODEL = `${URL_PREFIX}-qwen-3/${VERSION_TAG}/qwen-3-4B/
 const QWEN3_TOKENIZER = `${URL_PREFIX}-qwen-3/${VERSION_TAG}/tokenizer.json`;
 const QWEN3_TOKENIZER_CONFIG = `${URL_PREFIX}-qwen-3/${VERSION_TAG}/tokenizer_config.json`;
 
+// Qwen3's published generation_config.json recommends temperature=0.6 and
+// top_p=0.95. We propagate those to every Qwen3 preset so model quality is
+// reasonable out of the box; users can override via `configure()`.
+const QWEN3_GENERATION_CONFIG = {
+  temperature: 0.6,
+  topP: 0.95,
+} as const;
+
 /**
  * @category Models - LLM
  */
@@ -91,6 +99,7 @@ export const QWEN3_0_6B = {
   modelSource: QWEN3_0_6B_MODEL,
   tokenizerSource: QWEN3_TOKENIZER,
   tokenizerConfigSource: QWEN3_TOKENIZER_CONFIG,
+  generationConfig: QWEN3_GENERATION_CONFIG,
 } as const;
 
 /**
@@ -101,6 +110,7 @@ export const QWEN3_0_6B_QUANTIZED = {
   modelSource: QWEN3_0_6B_QUANTIZED_MODEL,
   tokenizerSource: QWEN3_TOKENIZER,
   tokenizerConfigSource: QWEN3_TOKENIZER_CONFIG,
+  generationConfig: QWEN3_GENERATION_CONFIG,
 } as const;
 
 /**
@@ -111,6 +121,7 @@ export const QWEN3_1_7B = {
   modelSource: QWEN3_1_7B_MODEL,
   tokenizerSource: QWEN3_TOKENIZER,
   tokenizerConfigSource: QWEN3_TOKENIZER_CONFIG,
+  generationConfig: QWEN3_GENERATION_CONFIG,
 } as const;
 
 /**
@@ -121,6 +132,7 @@ export const QWEN3_1_7B_QUANTIZED = {
   modelSource: QWEN3_1_7B_QUANTIZED_MODEL,
   tokenizerSource: QWEN3_TOKENIZER,
   tokenizerConfigSource: QWEN3_TOKENIZER_CONFIG,
+  generationConfig: QWEN3_GENERATION_CONFIG,
 } as const;
 
 /**
@@ -131,6 +143,7 @@ export const QWEN3_4B = {
   modelSource: QWEN3_4B_MODEL,
   tokenizerSource: QWEN3_TOKENIZER,
   tokenizerConfigSource: QWEN3_TOKENIZER_CONFIG,
+  generationConfig: QWEN3_GENERATION_CONFIG,
 } as const;
 
 /**
@@ -141,6 +154,7 @@ export const QWEN3_4B_QUANTIZED = {
   modelSource: QWEN3_4B_QUANTIZED_MODEL,
   tokenizerSource: QWEN3_TOKENIZER,
   tokenizerConfigSource: QWEN3_TOKENIZER_CONFIG,
+  generationConfig: QWEN3_GENERATION_CONFIG,
 } as const;
 
 // HAMMER 2.1
@@ -444,25 +458,52 @@ const LFM2_VL_450M_TOKENIZER_CONFIG = `${URL_PREFIX}-lfm-2.5/${VERSION_TAG}/lfm2
 /**
  * @category Models - VLM
  */
-export const LFM2_VL_1_6B_QUANTIZED = {
+// LiquidAI's LFM2-VL model card recommends the following sampling settings.
+// Without them the model often produces generic / repetitive responses.
+const LFM2_5_VL_GENERATION_CONFIG = {
+  temperature: 0.1,
+  minP: 0.15,
+  repetitionPenalty: 1.05,
+} as const;
+
+/**
+ * @category Models - VLM
+ */
+export const LFM2_5_VL_1_6B_QUANTIZED = {
   modelName: 'lfm2.5-vl-1.6b-quantized',
   capabilities: ['vision'],
   modelSource: LFM2_VL_1_6B_QUANTIZED_MODEL,
   tokenizerSource: LFM2_VL_1_6B_TOKENIZER,
   tokenizerConfigSource: LFM2_VL_1_6B_TOKENIZER_CONFIG,
+  generationConfig: LFM2_5_VL_GENERATION_CONFIG,
 } as const;
 
 /**
  * @category Models - VLM
  */
-export const LFM2_VL_450M_QUANTIZED = {
+export const LFM2_5_VL_450M_QUANTIZED = {
   modelName: 'lfm2.5-vl-450m-quantized',
   capabilities: ['vision'],
   modelSource: LFM2_VL_450M_QUANTIZED_MODEL,
   tokenizerSource: LFM2_VL_450M_TOKENIZER,
   tokenizerConfigSource: LFM2_VL_450M_TOKENIZER_CONFIG,
+  generationConfig: LFM2_5_VL_GENERATION_CONFIG,
 } as const;
 
+/**
+ * @deprecated Use `LFM2_5_VL_1_6B_QUANTIZED` instead — the model is from the
+ * LFM2.5 family. This alias will be removed in a future major release.
+ * @category Models - VLM
+ */
+export const LFM2_VL_1_6B_QUANTIZED = LFM2_5_VL_1_6B_QUANTIZED;
+
+/**
+ * @deprecated Use `LFM2_5_VL_450M_QUANTIZED` instead — the model is from the
+ * LFM2.5 family. This alias will be removed in a future major release.
+ * @category Models - VLM
+ */
+export const LFM2_VL_450M_QUANTIZED = LFM2_5_VL_450M_QUANTIZED;
+
 // Classification
 const EFFICIENTNET_V2_S_MODEL =
   Platform.OS === `ios`
@@ -1121,6 +1162,8 @@ export const MODEL_REGISTRY = {
     LFM2_5_350M_QUANTIZED,
     LFM2_5_1_2B_INSTRUCT,
     LFM2_5_1_2B_INSTRUCT_QUANTIZED,
+    LFM2_5_VL_1_6B_QUANTIZED,
+    LFM2_5_VL_450M_QUANTIZED,
     LFM2_VL_1_6B_QUANTIZED,
     LFM2_VL_450M_QUANTIZED,
     EFFICIENTNET_V2_S,
diff --git a/packages/react-native-executorch/src/controllers/LLMController.ts b/packages/react-native-executorch/src/controllers/LLMController.ts
index d9d5593ae3..696825eee3 100644
--- a/packages/react-native-executorch/src/controllers/LLMController.ts
+++ b/packages/react-native-executorch/src/controllers/LLMController.ts
@@ -76,12 +76,14 @@ export class LLMController {
     tokenizerSource,
     tokenizerConfigSource,
     capabilities,
+    defaultGenerationConfig,
     onDownloadProgressCallback,
   }: {
     modelSource: ResourceSource;
     tokenizerSource: ResourceSource;
     tokenizerConfigSource: ResourceSource;
     capabilities?: readonly LLMCapability[];
+    defaultGenerationConfig?: GenerationConfig;
     onDownloadProgressCallback?: (downloadProgress: number) => void;
   }) {
     // reset inner state when loading new model
@@ -130,6 +132,12 @@ export class LLMController {
         tokenizerPath,
         capabilities ?? []
       );
+      if (defaultGenerationConfig) {
+        // Apply model-specific recommended sampling defaults before flipping
+        // isReady so callers that react to it see the right config on first
+        // send. User-provided `configure()` calls still override these.
+        this.applyGenerationConfig(defaultGenerationConfig);
+      }
       this.isReadyCallback(true);
       this.onToken = (data: string) => {
         if (!data) {
@@ -166,28 +174,58 @@ export class LLMController {
     this.chatConfig = { ...DEFAULT_CHAT_CONFIG, ...chatConfig };
     this.toolsConfig = toolsConfig;
 
-    if (generationConfig?.outputTokenBatchSize) {
+    if (generationConfig) {
+      this.applyGenerationConfig(generationConfig);
+    }
+
+    // reset inner state when loading new configuration
+    this.messageHistoryCallback(this.chatConfig.initialMessageHistory);
+    this.isGeneratingCallback(false);
+  }
+
+  private applyGenerationConfig(generationConfig: GenerationConfig) {
+    if (generationConfig.outputTokenBatchSize) {
       this.nativeModule.setCountInterval(generationConfig.outputTokenBatchSize);
     }
-    if (generationConfig?.batchTimeInterval) {
+    if (generationConfig.batchTimeInterval) {
       this.nativeModule.setTimeInterval(generationConfig.batchTimeInterval);
     }
-    if (generationConfig?.temperature) {
+    if (generationConfig.temperature !== undefined) {
       this.nativeModule.setTemperature(generationConfig.temperature);
     }
-    if (generationConfig?.topp) {
-      if (generationConfig.topp < 0 || generationConfig.topp > 1) {
+    // `topp` is the legacy spelling kept for backwards compatibility — `topP`
+    // wins when both are set so callers migrating to the new name don't get
+    // surprised by stale values. Reading the deprecated alias is intentional.
+    const topP = generationConfig.topP ?? generationConfig.topp;
+    if (topP !== undefined) {
+      if (topP < 0 || topP > 1) {
         throw new RnExecutorchError(
           RnExecutorchErrorCode.InvalidConfig,
           'Top P has to be in range [0, 1]'
         );
       }
-      this.nativeModule.setTopp(generationConfig.topp);
+      this.nativeModule.setTopp(topP);
+    }
+    if (generationConfig.minP !== undefined) {
+      if (generationConfig.minP < 0 || generationConfig.minP > 1) {
+        throw new RnExecutorchError(
+          RnExecutorchErrorCode.InvalidConfig,
+          'Min P has to be in range [0, 1]'
+        );
+      }
+      this.nativeModule.setMinP(generationConfig.minP);
+    }
+    if (generationConfig.repetitionPenalty !== undefined) {
+      if (generationConfig.repetitionPenalty < 0) {
+        throw new RnExecutorchError(
+          RnExecutorchErrorCode.InvalidConfig,
+          'Repetition penalty must be non-negative'
+        );
+      }
+      this.nativeModule.setRepetitionPenalty(
+        generationConfig.repetitionPenalty
+      );
     }
-
-    // reset inner state when loading new configuration
-    this.messageHistoryCallback(this.chatConfig.initialMessageHistory);
-    this.isGeneratingCallback(false);
   }
 
   private getImageToken(): string {
diff --git a/packages/react-native-executorch/src/hooks/natural_language_processing/useLLM.ts b/packages/react-native-executorch/src/hooks/natural_language_processing/useLLM.ts
index feb7a931e3..027e237997 100644
--- a/packages/react-native-executorch/src/hooks/natural_language_processing/useLLM.ts
+++ b/packages/react-native-executorch/src/hooks/natural_language_processing/useLLM.ts
@@ -62,6 +62,7 @@ export function useLLM({
           tokenizerSource: model.tokenizerSource,
           tokenizerConfigSource: model.tokenizerConfigSource!,
           capabilities: model.capabilities,
+          defaultGenerationConfig: model.generationConfig,
           onDownloadProgressCallback: setDownloadProgress,
         });
       } catch (e) {
diff --git a/packages/react-native-executorch/src/types/llm.ts b/packages/react-native-executorch/src/types/llm.ts
index 3dbf54bb94..b40edf4dd7 100644
--- a/packages/react-native-executorch/src/types/llm.ts
+++ b/packages/react-native-executorch/src/types/llm.ts
@@ -87,6 +87,13 @@ export interface LLMProps {
      * Example: `['vision']` enables `sendMessage(text, { imagePath })`.
      */
     capabilities?: readonly LLMCapability[];
+    /**
+     * Recommended default generation settings, typically copied from the
+     * upstream `generation_config.json` or the model card. Applied automatically
+     * after the native module loads and before any user `configure()` call,
+     * so callers only need to override the values they want to change.
+     */
+    generationConfig?: GenerationConfig;
   };
   /**
    * Boolean that can prevent automatic model loading (and downloading the data if you load it for the first time) after running the hook.
@@ -248,6 +255,10 @@ export interface LLMConfig {
    * `temperature` - Scales output logits by the inverse of temperature. Controls the randomness / creativity of text generation.
    *
    * `topp` - Only samples from the smallest set of tokens whose cumulative probability exceeds topp.
+   *
+   * `minP` - Minimum probability threshold: tokens with prob < minP * max_prob are excluded. 0 disables filtering.
+   *
+   * `repetitionPenalty` - Multiplicative penalty applied to logits of recently generated tokens. Values > 1 discourage repetition. 1 disables the penalty.
    */
   generationConfig?: GenerationConfig;
 }
@@ -325,13 +336,20 @@ export interface ToolsConfig {
  * Object configuring generation settings.
  * @category Types
  * @property {number} [temperature] - Scales output logits by the inverse of temperature. Controls the randomness / creativity of text generation.
- * @property {number} [topp] - Only samples from the smallest set of tokens whose cumulative probability exceeds topp.
+ * @property {number} [topP] - Only samples from the smallest set of tokens whose cumulative probability exceeds topP.
+ * @property {number} [topp] - **Deprecated.** Use `topP` instead.
+ * @property {number} [minP] - Minimum probability threshold: tokens with prob < minP * max_prob are excluded. 0 disables filtering.
+ * @property {number} [repetitionPenalty] - Multiplicative penalty applied to logits of recently generated tokens. Values > 1 discourage repetition. 1 disables the penalty.
  * @property {number} [outputTokenBatchSize] - Soft upper limit on the number of tokens in each token batch (in certain cases there can be more tokens in given batch, i.e. when the batch would end with special emoji join character).
  * @property {number} [batchTimeInterval] - Upper limit on the time interval between consecutive token batches.
  */
 export interface GenerationConfig {
   temperature?: number;
+  topP?: number;
+  /** @deprecated Use `topP` instead. */
   topp?: number;
+  minP?: number;
+  repetitionPenalty?: number;
   outputTokenBatchSize?: number;
   batchTimeInterval?: number;
 }

From 78e7fd7940011210011f13e64bf6d27a1856cfd8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Mateusz=20S=C5=82uszniak?= <mateusz.sluszniak@swmansion.com>
Date: Tue, 28 Apr 2026 18:34:49 +0200
Subject: [PATCH 5/5] chore: Release v0.8.4

---
 packages/react-native-executorch/package.json | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/packages/react-native-executorch/package.json b/packages/react-native-executorch/package.json
index 2e76688f90..db229bcfaa 100644
--- a/packages/react-native-executorch/package.json
+++ b/packages/react-native-executorch/package.json
@@ -1,6 +1,6 @@
 {
   "name": "react-native-executorch",
-  "version": "0.8.3",
+  "version": "0.8.4",
   "description": "An easy way to run AI models in React Native with ExecuTorch",
   "source": "./src/index.ts",
   "main": "./lib/module/index.js",