Overview

1 Pull request merged by1 person

• [Fix] Fix gemma 3 conv template stop token

  #3191 mergedMar 27, 2025

2 Pull requests opened by1 person

• [Serving] Add Structural-Tag api to RequestResponseFormat

  #3187 openedMar 24, 2025

• [Serving] Support tool function calls under strict format constraints

  #3190 openedMar 26, 2025

2 Issues closed by2 people

• mlc_llm package error[Bug]

  #3180 closedMar 24, 2025

• [Bug] /mlc-llm/3rdparty/tvm/src/runtime/relax_vm/attn_utils.h:712:19: error: no member named 'cl' in namespace 'tvm::runtime'

  #3157 closedMar 21, 2025

1 Issue opened by1 person

• [Question] How to evaluate the accuracy of models???

  #3188 openedMar 24, 2025

6 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• [Bug] clang linker error upon running any model in Windows

  #3177 commented onMar 22, 2025 • 0 new comments

• [Feature Request] Support for returning log probabilities of both the prompt and the response tokens in the MLC-LLM API, similar to the functionality provided by OpenAI API.

  #2908 commented onMar 24, 2025 • 0 new comments

• [Bug] Broken for Intel Macs since v0.15 (or earlier)

  #3078 commented onMar 25, 2025 • 0 new comments

• [Bug] Mlc cli server gets stuck

  #3145 commented onMar 27, 2025 • 0 new comments

• [Bug] Binary was created using {relax.Executable} but a loader of that name is not registered. Available loaders are relax.VMExecutable, const_loader, static_library, metal. Perhaps you need to recompile with this runtime enabled.

  #3186 commented onMar 28, 2025 • 0 new comments

• Refactored random.h to have PhiloxRandomGenerator

  #3181 commented onMar 23, 2025 • 0 new comments