OptimalScale/LMFlow

View on GitHub

An Extensible Toolkit for Finetuning and Inference of Large Foundation Models. Large Models for All.

Stars: 8.5k
Forks: 822
Open beginner issues: 0
Indexed issues: 79

Dominant language: Python
License: Apache-2.0
Last GitHub push: Aug 10, 2026
Latest indexed: Sep 20, 2026
Contributing guide: Contributing guide
Code of conduct: Code of conduct
Beginner labels: No beginner labels indexed
PR merge metrics: No merged PRs in 30d

Beginner friendly Everything

Open All

79 open issues indexed

Use `dtype` instead of deprecated `torch_dtype` for transformers >= 4.56 Open

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

OptimalScale/LMFlow#977 · Opened Aug 28, 2026
Security: request for a private disclosure contact Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

OptimalScale/LMFlow#973 · Opened Jul 12, 2026
📋 Documentation Enhancement Suggestion Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

OptimalScale/LMFlow#969 · Opened May 11, 2026
[BUG] Incompatible Package Versions Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

OptimalScale/LMFlow#962 · Opened Jan 15, 2026
[BUG]Connecting to lmflow.org (lmflow.org)|107.23.182.175|:5000... failed: Connection timed out. Open

Difficulty 4/5 3-5 days Newbie friendliness 20/100

OptimalScale/LMFlow#954 · 1 comment · 4 reactions · Opened Nov 3, 2025
Don't understand LISA Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

OptimalScale/LMFlow#953 · 1 comment · Opened Aug 16, 2025
[BUG] Open

pending

Difficulty 3/5 1-2 days Newbie friendliness 25/100

OptimalScale/LMFlow#942 · Opened Apr 3, 2025
Pleaee @showgood880702 i need this dataset as text2text .json file to understand the structure correctly Open

Difficulty 5/5 Over a week Newbie friendliness 15/100

OptimalScale/LMFlow#941 · 1 comment · Opened Apr 1, 2025
Connection time out Open

Difficulty 2/5 1-3 hours Newbie friendliness 25/100

OptimalScale/LMFlow#940 · 2 comments · Opened Mar 22, 2025
[BUG]unexpected behavior on qwen 2.5 series templates Open

Difficulty 2/5 1-3 hours Newbie friendliness 45/100

OptimalScale/LMFlow#934 · 1 comment · Opened Feb 26, 2025
[BUG] Cannot run the raft_align with model Qwen2.5 Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

OptimalScale/LMFlow#932 · Opened Feb 14, 2025
How to train a multimodal model with LMFlow? Open

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

OptimalScale/LMFlow#929 · Opened Jan 6, 2025
What is the difference between LMFlow and LLaMA Factory Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

OptimalScale/LMFlow#923 · Opened Dec 20, 2024
[BUG] Do not use generation of hf model in inferencers Open

pending

Difficulty 3/5 1-2 days Newbie friendliness 45/100

OptimalScale/LMFlow#920 · 2 reactions · Opened Dec 2, 2024
Conversation_template Open

enhancement

Difficulty 3/5 1-2 days Newbie friendliness 30/100

OptimalScale/LMFlow#917 · 4 comments · Opened Nov 6, 2024
Can LMFlow benchmark evaluate my finetuned model on user-defined dataset？ Open

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

OptimalScale/LMFlow#916 · Opened Nov 6, 2024
Pre-training Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

OptimalScale/LMFlow#910 · Opened Oct 17, 2024
Does it support qwen2.5? Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

OptimalScale/LMFlow#907 · 1 reaction · Opened Oct 12, 2024
How to implement weight decay towards the pre-trained model? Open

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

OptimalScale/LMFlow#906 · Opened Oct 3, 2024
Some question on code of LISA Open

Difficulty 4/5 3-5 days Newbie friendliness 15/100

OptimalScale/LMFlow#897 · 1 comment · Opened Sep 19, 2024
Does it support qwen2? Open

Difficulty 5/5 Over a week Newbie friendliness 15/100

OptimalScale/LMFlow#896 · 2 comments · Opened Aug 26, 2024
Specific methods about lisa Open

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

OptimalScale/LMFlow#895 · 1 comment · Opened Aug 23, 2024
Can lisa training support MiniCPM-V? Open

Difficulty 5/5 Over a week Newbie friendliness 15/100

OptimalScale/LMFlow#894 · 1 comment · Opened Aug 23, 2024
How can I use other optimizers in LISA? Open

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

OptimalScale/LMFlow#893 · 1 comment · Opened Aug 22, 2024
Is there any Lisa that supports llama3.1 yet? Open

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

OptimalScale/LMFlow#890 · 1 comment · Opened Aug 12, 2024
Problems encountered during speculative decoding execution Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

OptimalScale/LMFlow#888 · 1 comment · Opened Jul 31, 2024
[BUG]使用远程服务器运行chatbot_gradio.py时意外中断连接 Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

OptimalScale/LMFlow#886 · 2 comments · Opened Jul 24, 2024
[BUG] EvaluatorArguments, Open

Difficulty 2/5 1-3 hours Newbie friendliness 35/100

OptimalScale/LMFlow#885 · 1 comment · Opened Jul 23, 2024
[New Feature] Is Mixtral supported? Open

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

OptimalScale/LMFlow#879 · 1 comment · Opened Jul 8, 2024
DPO+ZeRO train error Open

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

OptimalScale/LMFlow#870 · 2 comments · Opened Jun 27, 2024
RAFT Issues Open

Difficulty 4/5 3-5 days Newbie friendliness 20/100

OptimalScale/LMFlow#868 · 2 comments · Opened Jun 25, 2024
Full parameter fine-tuning bugs Open

Difficulty 5/5 Over a week Newbie friendliness 18/100

OptimalScale/LMFlow#864 · 3 comments · Opened Jun 20, 2024
[Roadmap] LMFlow Roadmap Open

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

OptimalScale/LMFlow#862 · 2 comments · 2 reactions · Opened Jun 19, 2024
[BUG] The text cannot be generated successfully during the Raft step Open

pending

Difficulty 4/5 3-5 days Newbie friendliness 25/100

OptimalScale/LMFlow#861 · 1 comment · Opened Jun 19, 2024
multi-gpu full para train error Open

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

OptimalScale/LMFlow#855 · 2 comments · Opened Jun 12, 2024
Full parameter fine-tuning cannot be trained Open

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

OptimalScale/LMFlow#842 · 1 comment · Opened May 22, 2024
Training was successful on a single card 4090GPU, but an error was reported on a 3*4090GPU. why Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

OptimalScale/LMFlow#841 · 1 comment · Opened May 22, 2024
[BUG]when map the dataset, i set the num_proc = 2 or 4, it will make mistakes. Open

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

OptimalScale/LMFlow#840 · 8 comments · Opened May 22, 2024
Discussion about LISA Open

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

OptimalScale/LMFlow#833 · 2 comments · Opened May 20, 2024
Weird Loss Curve Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

OptimalScale/LMFlow#831 · 1 comment · Opened May 17, 2024
[BUG] Can Lisa be used for chatglm3 with lmflow? Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

OptimalScale/LMFlow#825 · 1 comment · Opened May 14, 2024
Add 'validation_split_percentage' and 'evaluation_strategy' parameters for Trainers Open

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

OptimalScale/LMFlow#823 · 1 comment · Opened May 12, 2024
Fine-Tuning Crashes for no reason when Eight GPU cards are used. Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

OptimalScale/LMFlow#816 · 4 comments · Opened May 8, 2024
[BUG] did not output the eval results at all. Open

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

OptimalScale/LMFlow#815 · 3 comments · Opened May 8, 2024
Weird Loss with LISA Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

OptimalScale/LMFlow#806 · 1 comment · Opened May 4, 2024
Causal LM finetuning Open

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

OptimalScale/LMFlow#786 · 3 comments · Opened Apr 23, 2024
Does it support llama3? Open

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

OptimalScale/LMFlow#785 · 5 comments · Opened Apr 23, 2024
Memory problem of Lisa finetuning Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

OptimalScale/LMFlow#778 · 6 comments · Opened Apr 21, 2024
[New Feature] Could someone share the finetuned diffusion model which is good at 256x256 resolution? Open

Difficulty 5/5 Over a week Newbie friendliness 15/100

OptimalScale/LMFlow#777 · Opened Apr 20, 2024
About using multiple GPUs to do lisa fine-tuning Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

OptimalScale/LMFlow#774 · 4 comments · Opened Apr 19, 2024

OptimalScale/LMFlow

Get new issues in your inbox