Skip to content

Issues: mlc-ai/mlc-llm

Project Tracking
#647 opened Aug 2, 2023 by tqchen
Open
Model Request Tracking
#1042 opened Oct 9, 2023 by CharlieFRuan
Open 4
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

218 Open 1,250 Closed
218 Open 1,250 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

[Bug] bug Confirmed bugs
#3143 opened Feb 20, 2025 by leezear2022
[Question] mlc-llm server cannot return correct logprobs question Question about the usage
#3142 opened Feb 19, 2025 by kunxiongzhu
1
[Question] how to use function call question Question about the usage
#3141 opened Feb 19, 2025 by tebie6
[Model Request] GLINER for entity recognisition new-models
#3139 opened Feb 17, 2025 by manasaniprashanth
[Bug] Gemma 2 models fail due to errors in tokenizer bug Confirmed bugs
#3138 opened Feb 17, 2025 by julioasotodv
[Question] I followed the instructions to build for Orange Pi, but it seems outdated (ChatModule) question Question about the usage
#3134 opened Feb 16, 2025 by LivingLinux
[Bug] Softmax op is very slow bug Confirmed bugs
#3132 opened Feb 13, 2025 by gesanqiu
[Question] While waiting for the model's response on an Android phone, performing other operations may cause the phone to become unresponsive or reboot. question Question about the usage
#3131 opened Feb 13, 2025 by yangshgetui
2
[Bug] mlc-llm server cannot return correct logprobs bug Confirmed bugs
#3130 opened Feb 13, 2025 by kunxiongzhu
1
[Bug] Is it compiling? CUDA 12.8 bug Confirmed bugs
#3129 opened Feb 12, 2025 by johnnynunez
[Bug] Mistral-Nemo-Instruct-2407 The results were confused bug Confirmed bugs
#3120 opened Feb 7, 2025 by fierceX
Very slow time to first token on ROCM question Question about the usage
#3119 opened Feb 5, 2025 by Jyers
3
[Bug] Android app does not take input; 'user 'role' is not defined' error bug Confirmed bugs
#3117 opened Feb 4, 2025 by afsara-ben
5
How to stop a stream? question Question about the usage
#3113 opened Jan 30, 2025 by hpssjellis
7
[Question] Deepseek R1 Distill Qwen 1.5B converted models have very large VRAM requirement. question Question about the usage
#3112 opened Jan 28, 2025 by bhushangawde
10
[Bug] Compiling the MLC from source is failed (cuda_fp8.h) bug Confirmed bugs
#3111 opened Jan 26, 2025 by wwt02
4
[Bug] Failed to compile model on aarch64 platform with cuda12.8 bug Confirmed bugs
#3110 opened Jan 26, 2025 by XiChen0415
5
[Question] Difference between the quantization methods of other LLM engines. question Question about the usage
#3107 opened Jan 23, 2025 by BrandonLee0626
[Model Request] minicpm-o-2.6 model new-models
#3100 opened Jan 21, 2025 by timber-vine
3
[Question]In the output results of attention_with_fused_qkv funcs, some slice accuracies are abnormal question Question about the usage
#3093 opened Jan 17, 2025 by ifndefendif
6
[Question] Android App Crash question Question about the usage
#3091 opened Jan 16, 2025 by mhollis1980
1
[Question] semantic description of different quantization methods question Question about the usage
#3088 opened Jan 9, 2025 by phgcha
[Bug] Broken for Intel Macs since v0.15 (or earlier) bug Confirmed bugs
#3078 opened Dec 31, 2024 by zxcat
6
[Bug] cohere model(aya) doesn't seem to produce the correct output bug Confirmed bugs
#3073 opened Dec 21, 2024 by jhlee525
[Feature Request] Provide a C++ API feature request New feature or request
#3066 opened Dec 16, 2024 by tranlm
2
ProTip! Follow long discussions with comments:>50.