Skip to content

Issues: unslothai/unsloth

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

657 Open 619 Closed
657 Open 619 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

Allow vLLM on 2nd GPU for GRPO training
#1635 opened Feb 7, 2025 by kallewoof
Unable to Patch GRPO in Kaggle
#1634 opened Feb 7, 2025 by IsNoobgrammer
1
I can not run unsloth/qwen2.5-7b-bnb-4bit because it is not supported in my current Unsloth version
#1633 opened Feb 7, 2025 by NguyenTrinh3008
1
Model not supported - needed for research paper - b1ade-1b-bf16
#1630 opened Feb 7, 2025 by w601sxs
ValueError: Some modules are dispatched on the CPU or the disk
#1629 opened Feb 7, 2025 by Sweaterdog
TOKENIZER DOESN'T WORK! PLEASE HELP!
#1628 opened Feb 7, 2025 by yukiarimo
Error when loading Almawave/Velvet-14B tokenizer
#1626 opened Feb 6, 2025 by dtdxdydz
GRPOTrainer crashes with unsloth
#1624 opened Feb 6, 2025 by ymcki
13
Got infinite repetition results when running the DeepSeek-R1-UD-IQ1_S (1.5bit dynamic quantized version) on Nvidia A800 GPU by llama.cpp
#1622 opened Feb 6, 2025 by Jisencc
5
Orpo trainer is reporting loss without batchsize/gradiend accumulation taken into account currently fixing Am fixing now!
#1619 opened Feb 6, 2025 by Nazzaroth2
1
ModuleNotFoundError: No module named 'torch' - it's realy, module is installed
#1616 opened Feb 4, 2025 by lexasub
2
[Docs Improvement] Improve documentation on how to export model from Colab feature request Feature request pending on roadmap good first issue help wanted Help from the OSS community wanted!
#1615 opened Feb 4, 2025 by gaspardc-met
1
[Question/Docs] How to use "Raw Corpus" text for the data prep
#1614 opened Feb 4, 2025 by gaspardc-met
Qwen2.5-VL-3B 4Bit train
#1613 opened Feb 4, 2025 by wpprpoew
5
Context length for DPO on a 7b model
#1612 opened Feb 4, 2025 by gmonair
2
PatchDPOTrainer bug
#1611 opened Feb 4, 2025 by sihaogu32
1
llama.cpp GGUF breaks
#1610 opened Feb 4, 2025 by Eleanorkong
Weights only load failed when target_modules contain "embed_tokens", "lm_head"
#1606 opened Feb 3, 2025 by anhnh2002
Error running Mistral small 2501 on vllm
#1605 opened Feb 3, 2025 by thesillystudent
1
TypeError: must be called with a dataclass type or instance
#1604 opened Feb 2, 2025 by azzedineA
1
Using llama-cpp-python to run quantized R1
#1603 opened Feb 2, 2025 by ZN1010
Should quantization_config section of merged_16bit model be removed, otherwise it is causing errors
#1601 opened Feb 2, 2025 by fzyzcjy
unsloth/Mistral-Small-24B-Instruct-2501 RuntimeError: The size of tensor a (128) must match the size of tensor b (160) at non-singleton dimension 3
#1600 opened Feb 1, 2025 by ichrnkv
7
Mistral Small: ValueError: Blockwise quantization only supports 16/32-bit floats, but got torch.uint8
#1599 opened Jan 31, 2025 by DaddyCodesAlot
3
[REQUEST] DeepSeek-R1-Distill-Qwen-32B-GGUF 1.58-bit version possible ?
#1594 opened Jan 30, 2025 by Greatz08
2
ProTip! Exclude everything labeled bug with -label:bug.