fix: crash when sending image under 2x2 pixels#21711
Merged
ngxson merged 3 commits intoggml-org:masterfrom Apr 12, 2026
Merged
Conversation
ngxson
approved these changes
Apr 11, 2026
Contributor
pwilkin
approved these changes
Apr 11, 2026
crodjer
added a commit
to crodjer/llama.cpp
that referenced
this pull request
Apr 13, 2026
* origin/master: webui: MCP Diagnostics improvements (ggml-org#21803) Remove extra conditional check on debug mode. (ggml-org#21798) sycl: disable Q1_0 in backend and cleanup unused variables (ggml-org#21807) mtmd: fix crash when sending image under 2x2 pixels (ggml-org#21711) mtmd: qwen3 audio support (qwen3-omni and qwen3-asr) (ggml-org#19441) convert : force f16 or f32 on step3-vl conv weights (ggml-org#21646) mtmd: add gemma 4 test (vision + audio) [no ci] (ggml-org#21806) mtmd: add Gemma 4 audio conformer encoder support (ggml-org#21421) fix: Proper messages rendering for "Show raw output" (ggml-org#21672) docs: add guide on how to add multimodal support (ggml-org#21778)
HermestoAizales
pushed a commit
to HermestoAizales/llama.cpp
that referenced
this pull request
Apr 13, 2026
ArberSephirotheca
pushed a commit
to ArberSephirotheca/llama.cpp
that referenced
this pull request
Apr 21, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Overview
GGML_ASSERT(src.nx >= 2 && src.ny >= 2); will crash llama.cpp when processing very small images. Fix was implemented to handle 1x1 inputs safely by updating the interpolation math and clamping pixel lookups, preventing out-of-bounds memory errors while keeping the pipeline stable.
Code was succesfully tested in production, llama-server is running with no crashes.
Additional information
Fixes #21420
Requirements