Owner

mlc-ai

18 indexed repositories · View on GitHub

mlc-llm
Universal LLM Deployment Engine with ML Compilation

Python · 23170 stars
web-llm
High-performance In-browser LLM Inference Engine

TypeScript · 19151 stars
web-stable-diffusion
Bringing stable diffusion models to web browsers. Everything runs inside the browser with no server support.

Jupyter Notebook · 3725 stars
xgrammar
Fast, Flexible and Portable Structured Generation

C++ · 1896 stars
modern-gpu-programming-for-mlsys
A tutorial on modern GPU programming for machine learning systems

HTML · 1290 stars
web-llm-chat
Chat with AI large language models running natively in your browser. Enjoy private, server-free, seamless AI conversations.

TypeScript · 1106 stars
mlc-zh

Python · 639 stars
tokenizers-cpp
Universal cross-platform tokenizers binding to HF and sentencepiece

C++ · 512 stars
mlc-en

Python · 462 stars
pith-train
Compact and Agent-Native MoE Training System

Python · 352 stars
binary-mlc-llm-libs

Python · 330 stars
notebooks

Jupyter Notebook · 229 stars
relax

Python · 175 stars
llm-perf-bench

Python · 122 stars
web-llm-assistant
AI Assistant running within your browser.

TypeScript · 85 stars
mlc-python

C++ · 36 stars
web-agent-interface
An interface for AI Agents to interact with various web platforms.

TypeScript · 19 stars
package

Python · 14 stars

Beginner friendly Everything

Open All

[Feature Request] configurable download threads Open

Difficulty 3/5 1-2 days Newbie friendliness 52/100

mlc-ai/web-llm#857 · Opened Sep 12, 2026
License and attribution clarification for Qwen2.5-0.5B WebGPU model library (v0_2_84) Open

Difficulty 4/5 3-5 days Newbie friendliness 45/100

mlc-ai/binary-mlc-llm-libs#166 · Opened Sep 11, 2026
[First-time contributor] Looking for beginner-friendly issues to contribute Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

mlc-ai/mlc-llm#3547 · Opened Sep 8, 2026
Exposing a compatible interface for Chrome Built-in AI (Prompt API) / Polyfill Open

Difficulty 5/5 Over a week Newbie friendliness 28/100

mlc-ai/web-llm#853 · 5 comments · Opened Sep 3, 2026
[Bug] `fill_next_token_bitmask` latency grows linearly with the number of accepted tokens (0.2.6rc1/rc2 regression; fine on 0.2.3) Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 72/100

mlc-ai/xgrammar#873 · 5 comments · Opened Sep 2, 2026
[Feature Request] Open

feature request

Difficulty 5/5 Over a week Newbie friendliness 25/100

mlc-ai/mlc-llm#3544 · Opened Sep 2, 2026
Ninja and Cmake should not be in pyproject's build-system.requires Open

Difficulty 2/5 1-3 hours Newbie friendliness 78/100

mlc-ai/package#114 · Opened Aug 31, 2026
[Bug] ToString() and JSON serialization silently drop bytes after an embedded NUL Open

Difficulty 3/5 1-2 days Newbie friendliness 72/100

mlc-ai/xgrammar#866 · 1 comment · Opened Aug 27, 2026
[Bug] JSON schema converter silently drops sibling keywords (properties, additionalProperties, ...) when anyOf/oneOf/allOf is present at the same node Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 52/100

mlc-ai/xgrammar#858 · Opened Aug 22, 2026
Downloadable APKs for newer versions Open

feature request

Difficulty 5/5 Over a week Newbie friendliness 35/100

mlc-ai/mlc-llm#3537 · 1 comment · Opened Aug 21, 2026
[Bug] macOS ARM64 mlc-llm-nightly-cpu pip wheel crashes entire Python process via silent exit(0) during import mlc_llm Open

bug

Difficulty 5/5 Over a week Newbie friendliness 38/100

mlc-ai/mlc-llm#3536 · Opened Aug 21, 2026
[Bug] [Perf] String `maxLength` in a JSON schema makes `fill_next_token_bitmask` ~300x slower Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 48/100

mlc-ai/xgrammar#852 · 1 comment · 1 reaction · Opened Aug 20, 2026
disable_any_whitespace on schemas with additionalProperties: model dumps its answer into an empty string key, quality collapses Open

Difficulty 2/5 1-3 hours Newbie friendliness 62/100

mlc-ai/xgrammar#851 · Opened Aug 19, 2026
Feature request Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

mlc-ai/web-llm#847 · Opened Aug 17, 2026
Negated character class over non-ASCII ranges silently fails to enforce (clamped to 127) Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 55/100

mlc-ai/xgrammar#848 · Opened Aug 17, 2026
Notice: version 0.1.28 removed from PyPI Open

notice

Difficulty 1/5 Under an hour Newbie friendliness 20/100

mlc-ai/xgrammar#845 · Opened Aug 15, 2026
[Bug] draft-07 additionalItems is silently ignored (true and schema forms reject valid documents) Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 35/100

mlc-ai/xgrammar#840 · 2 comments · Opened Aug 14, 2026
[Bug] pattern/patternProperties use full-string matching instead of Draft 2020-12 unanchored search Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 45/100

mlc-ai/xgrammar#838 · 1 comment · Opened Aug 14, 2026
[Bug] type:array without items/prefixItems compiles to an only-empty-array grammar when any keyword is present Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 58/100

mlc-ai/xgrammar#833 · 1 comment · Opened Aug 14, 2026
[Feature] Allow callers to declare special/control token ids for TokenizerInfo Open

feature-request

Difficulty 4/5 3-5 days Newbie friendliness 50/100

mlc-ai/xgrammar#832 · Opened Aug 13, 2026
[Bug] With additionalProperties, a declared property named out of schema order can only be closed by escaping the quote Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 48/100

mlc-ai/xgrammar#831 · 3 comments · Opened Aug 13, 2026
[Bug] shapeCache LRU eviction disposes in-use ShapeTuples → "Object has already been disposed" → GPU device hang (regression introduced in 0.2.83) Open

mlc-ai/web-llm#844 · 1 comment · 1 assignee · Opened Aug 6, 2026
[Bug] Unbounded string minLength > 128 causes a fill_next_token_bitmask latency cliff Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 48/100

mlc-ai/xgrammar#805 · 1 comment · Opened Aug 6, 2026
[Bug] No matching distribution found for xgrammar==0.1.16 Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 42/100

mlc-ai/xgrammar#804 · 3 comments · Opened Aug 5, 2026
[Bug] XML tool calling loops forever on no-argument tools Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 72/100

mlc-ai/xgrammar#802 · 4 comments · Opened Aug 4, 2026
JSON Schema maxLength/minLength/pattern on strings drops the escape branch — \n, \", \\, \uXXXX become unemittable Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 57/100

mlc-ai/xgrammar#800 · 2 comments · Opened Aug 3, 2026
Proposal: Time to the best-performing correct configuration and agent skill for finding the best-performing correct training configuration Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

mlc-ai/pith-train#81 · Opened Jul 30, 2026
Is there any plan to support VLM? Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

mlc-ai/pith-train#80 · Opened Jul 30, 2026
pattern/format combined with minLength/maxLength silently drops the length bounds Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 48/100

mlc-ai/xgrammar#749 · 1 comment · Opened Jul 28, 2026
[Feature Request] Speculative decoding (draft model + batched verification) in the web runtime Open

mlc-ai/web-llm#840 · 1 comment · 1 assignee · Opened Jul 25, 2026
[Bug] AttributeError: module 'tvm.ir._ffi_api' has no attribute 'RegisterOpAttr' when running mlc_llm serve -h Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 45/100

mlc-ai/mlc-llm#3528 · 2 comments · Opened Jul 23, 2026
[Bug] MLC-LLM CLI Crashes with malloc Error on macOS after Source Build Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 45/100

mlc-ai/mlc-llm#3526 · 1 comment · Opened Jul 23, 2026
[Doc] Build from source instructions incomplete - missing Python environment setup and library path configuration Open

documentation

Difficulty 3/5 1-2 days Newbie friendliness 68/100

mlc-ai/mlc-llm#3525 · Opened Jul 23, 2026
does this still work Open

Difficulty 3/5 1-2 days Newbie friendliness 28/100

mlc-ai/web-stable-diffusion#69 · Opened Jul 23, 2026
[Bug] mlc-ai-nightly-cpu==0.26.dev203 cannot install due to apache-tvm-ffi>=0.1.13 version conflict on PyPI Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 56/100

mlc-ai/mlc-llm#3520 · Opened Jul 22, 2026
Question: Qwen3-30B-A3B throughput difference between PithTrain EP4/PP2 and TorchTitan FSDP8 on 8×B200 Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

mlc-ai/pith-train#79 · 7 comments · 1 reaction · Opened Jul 22, 2026
[Bug] installation 26.0.1 Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 35/100

mlc-ai/mlc-llm#3517 · Opened Jul 19, 2026
[Bug] M4 OSX 15.7.4 Installation Issue Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 42/100

mlc-ai/mlc-llm#3516 · Opened Jul 19, 2026
[Bug] Stop token that is also a grammar terminal can never be accepted Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 55/100

mlc-ai/xgrammar#701 · 2 comments · Opened Jul 18, 2026
get_generation_config mutates the request's stop list in place when stop is passed as a list Open

Difficulty 2/5 1-3 hours Newbie friendliness 72/100

mlc-ai/mlc-llm#3512 · 1 comment · Opened Jul 12, 2026
Publish a @mlc-ai/web-xgrammar build that includes the UTF-8 character class fix (#491) Open

feature-request

Difficulty 4/5 3-5 days Newbie friendliness 52/100

mlc-ai/xgrammar#696 · Opened Jul 12, 2026
[Bug] libtvm.so not found + GlobalFunctionTable duplicate registration on both nightly-cpu and nightly-cu128 (mlc-llm/mlc-ai desync) Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 38/100

mlc-ai/mlc-llm#3508 · Opened Jul 3, 2026
[Bug] Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 68/100

mlc-ai/mlc-llm#3506 · 2 comments · 1 reaction · Opened Jul 1, 2026
[Feature] Native support in xgrammar for transformers beam search Open

feature-request

Difficulty 4/5 3-5 days Newbie friendliness 45/100

mlc-ai/xgrammar#681 · Opened Jun 30, 2026
[Perf-Issue] cold-compile is slow for tool-calling tags with many tools Open

question

Difficulty 5/5 Over a week Newbie friendliness 38/100

mlc-ai/xgrammar#680 · 3 comments · Opened Jun 29, 2026
Suggestion: add coverage of multi-GPU kernels (at least single-node) Open

Difficulty 5/5 Over a week Newbie friendliness 38/100

mlc-ai/modern-gpu-programming-for-mlsys#131 · 1 comment · Opened Jun 27, 2026
feat: optional persistent on-disk cache for compiled grammars Open

feature-request

Difficulty 5/5 Over a week Newbie friendliness 35/100

mlc-ai/xgrammar#675 · Opened Jun 25, 2026
[Bug] WebLLM Engine Initialization Fails on Qualcomm Adreno GPUs with VK_ERROR_DEVICE_LOST Open

bug

Difficulty 5/5 Over a week Newbie friendliness 35/100

mlc-ai/mlc-llm#3502 · Opened Jun 21, 2026
[Bug] WebLLM Engine Initialization Fails on Qualcomm Adreno GPUs with VK_ERROR_DEVICE_LOST Open

Difficulty 5/5 Over a week Newbie friendliness 30/100

mlc-ai/web-llm#836 · 1 comment · Opened Jun 21, 2026
[Bug] mlc-llm-nightly-cpu mlc-ai-nightly-cpu Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 35/100

mlc-ai/mlc-llm#3500 · 3 comments · Opened Jun 9, 2026

Showing the newest 100

This page lists what was indexed most recently. Advanced filter has the whole inventory, narrowed by language, difficulty and how long a task takes.

Open advanced filter

mlc-ai

Showing the newest 100

Get new issues in your inbox