Scaffold Dhamaka: browser-native LLM stack with cross-site model hub#1
Merged
Merged
Conversation
Build out the full Dhamaka architecture end-to-end so the whole developer surface runs today, even while the WASM inference engine is still under construction. Packages: - @dhamaka/hub cross-site model cache (iframe + IndexedDB + postMessage) - @dhamaka/runtime Engine interface, MockEngine (working), WasmEngine (loader) - dhamaka public SDK: Dhamaka.load / complete / stream / chat / evict - @dhamaka/playground zero-dep dev server + live demo UI with telemetry Also: canonical manifest with the default dhamaka-micro model (SmolLM2-360M-Instruct Q4) and the planned registry variants, plus a sexy ASCII-art README documenting the architecture and the honest story on browser storage partitioning. The SDK talks to the hub over a typed postMessage protocol with progress events, zero-copy ArrayBuffer transfer, SHA-256 integrity checks, and a graceful per-origin fallback when the iframe isn't reachable. The OpenAI /v1/chat/completions shim is included for drop-in compatibility.
protosphinx
added a commit
that referenced
this pull request
Apr 11, 2026
Scaffold Dhamaka: browser-native LLM stack with cross-site model hub
protosphinx
added a commit
that referenced
this pull request
May 24, 2026
Scaffold Dhamaka: browser-native LLM stack with cross-site model hub
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Build out the full Dhamaka architecture end-to-end so the whole developer surface runs today, even while the WASM inference engine is still under construction.
Packages:
Also: canonical manifest with the default dhamaka-micro model (SmolLM2-360M-Instruct Q4) and the planned registry variants, plus a sexy ASCII-art README documenting the architecture and the honest story on browser storage partitioning.
The SDK talks to the hub over a typed postMessage protocol with progress events, zero-copy ArrayBuffer transfer, SHA-256 integrity checks, and a graceful per-origin fallback when the iframe isn't reachable. The OpenAI /v1/chat/completions shim is included for drop-in compatibility.