Skip to content

Conversation

@CharlieFRuan
Copy link
Member

Change

  • The only change is [Model] Support Qwen3 models with enable_thinking field #686, which
    • Add prebuilt models:
      • Qwen3-0.6B: q0f16, q0f32, q4f16_1, q4f32_1
      • Other Qwen3: {1.7B, 4B, 8B} x {q4f16_1, q4f32_1}
    • Support extra_body: {enable_thinking: false} for qwen3 models to toggle thinking
      • See examples/qwen3 for more on Qwen3 usage
    • Also bumped web-tokenizers package to 0.1.6 to resolve rust-related issues

TVMjs

  • No change, version 0.18.0-dev2 just like 0.2.71

@CharlieFRuan CharlieFRuan merged commit d8b25fe into mlc-ai:main May 5, 2025
1 check passed
atebites-hub pushed a commit to atebites-hub/web-llm that referenced this pull request Oct 4, 2025
### Change
- The only change is mlc-ai#686, which
  - Add prebuilt models:
    - Qwen3-0.6B: `q0f16, q0f32, q4f16_1, q4f32_1`
    - Other Qwen3: `{1.7B, 4B, 8B} x {q4f16_1, q4f32_1}`
- Support `extra_body: {enable_thinking: false}` for qwen3 models to
toggle thinking
    - See `examples/qwen3` for more on Qwen3 usage
- Also bumped `web-tokenizers` package to `0.1.6` to resolve
rust-related issues


### TVMjs
- No change, version `0.18.0-dev2` just like 0.2.71
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant