Starstruk Logo MarkStarstruk

v0.2.81

amirivojdan/shekarv0.2.81Feb 18, 2026by akaashrp

AI Summary

This release features a major engine and API overhaul, including multi-model loading and OpenAI-compatible APIs, alongside expanded support for various LLMs and WebGPU optimizations.

Key Highlights

  • Refactored ChatModule into Engine/MLCEngine with improved worker lifecycle and concurrency handling.
  • Added OpenAI API compatibility including chat/completions, function calling, and embeddings support.
  • Expanded prebuilt model support to include Llama 3.2, Qwen 3, Gemma 2, and vision models.
  • Improved WebGPU performance and reliability with better OOM handling and IndexedDB caching.
  • Integrated XGrammar for JSON-schema and grammar-constrained generation.

New Features

  • OpenAI API mirroring with stateful options and embeddings support.
  • XGrammar integration for structured generation.
  • TVM-FFI integration refactoring for compatibility.
  • IndexedDB caching support for runtime and prebuilt versions.
  • New RAG/doc-chat examples and updated Chrome extension demos.

Full Release Notes

* Engine and API overhaul: ChatModule refactored into Engine/MLCEngine, consolidated constructor/reload behavior, multi-model loading, better worker lifecycle, concurrency handling
* OpenAI API: mirror chat/completions APIs, stateful options, function calling and embeddings support
* Conversation templates: unified conversation template schema with custom templates
* Expanded prebuilt model support: added support for more models (Llama 2/3/3.1/3.2, Mistral variants, Gemma 2, Qwen2/2.5/3, Phi family, including vision)
* Runtime and caching: WebGPU performance/reliability improvements (more GPU-side kernels, better OOM/deviceLost handling), wasm/prebuilt versioning updates, support IndexedDB caching
* XGrammar integration: JSON-schema/grammar-constrained generation, XGrammar structural tag
* TVM-FFI integration: refactor for compatibility with more recent TVM commits and TVM FFI 
* Examples: ServiceWorkerEngine + updated Chrome extension demos, new RAG/doc-chat examples, tool calls via structural tag
* CI: GitHub actions for linting and pre-commit hooks 

**Full Changelog**: https://github.com/mlc-ai/web-llm/compare/v0.2.0...v0.2.81