Skip to content

Issues: bitsandbytes-foundation/bitsandbytes

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

184 Open 942 Closed
184 Open 942 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

RuntimeError: invalid numeric default value in 0.45.5
#1588 opened Apr 9, 2025 by ianmage
lib.cquantize_blockwise_fp32 mutates absmax
#1587 opened Apr 9, 2025 by cyr0930
1
Not working with Windows Triton 3.2? Windows
#1583 opened Apr 3, 2025 by deadman3000
@matthewdouglas
1
eliminate most high impact graph breaks
#1579 opened Mar 27, 2025 by Titus-von-Koeller v0.46.0
[spike] prototype blog post examples to identify most impactful graph breaks from there
#1578 opened Mar 27, 2025 by Titus-von-Koeller
1 task
3
Does it work for gemma3 4b likely not a BNB issue proposing to close
#1572 opened Mar 25, 2025 by Oxi84
@matthewdouglas
3
All input tensors need to be on the same GPU, but found some tensors to not be on a GPU Windows
#1571 opened Mar 24, 2025 by SiRMHS
memory issue, prepare_model_for_kbit_training question Further information is requested
#1569 opened Mar 19, 2025 by wenzhifang
CUDA Setup failed despite GPU being available. what's wrrong? CUDA Setup Windows
#1568 opened Mar 17, 2025 by yoogon20
@matthewdouglas
6
CUDA Setup failed despite GPU being available. CUDA Setup
#1559 opened Mar 8, 2025 by dvShilenko
3
[chore] fix continuous release not working correctly chore
#1558 opened Mar 6, 2025 by Titus-von-Koeller
3 tasks
@Titus-von-Koeller
[spike] evaluate + prototype interaction of unified memory abstraction with custom_ops Optimizers Issues or feature requests relating to optimizers spike
#1556 opened Mar 5, 2025 by Titus-von-Koeller
3 tasks
@Titus-von-Koeller
Increased VRAM consumption when coupled with DDP
#1555 opened Mar 4, 2025 by TParcollet
Custom Ops: Unit Tests cross-platform high priority (first issues that will be worked on)
#1553 opened Mar 3, 2025 by matthewdouglas
5 tasks
v0.46.0
@matthewdouglas
[AMD ROCm] NotImplementedError: bnb_4bit_use_double_quant is not supported yet for CPU/XPU ROCm
#1551 opened Mar 2, 2025 by Looong01
Improve stack trace when C library does not load contributions-welcome We welcome contributions to fix this issue! Low Risk Risk of bugs in transformers and other libraries medium priority (will be worked on after all high priority issues)
#1548 opened Feb 28, 2025 by matthewdouglas v0.47.0
non_sign_bits miscalculation in the function 'create_dynamic_map' in bitsandbytes/functional.py high priority (first issues that will be worked on) Low Risk Risk of bugs in transformers and other libraries
#1541 opened Feb 26, 2025 by Talel-bm
@matthewdouglas
1
Wrong result 8bit blockwise quantization over float16 bug Something isn't working high priority (first issues that will be worked on) Low Risk Risk of bugs in transformers and other libraries x64 CPU
#1540 opened Feb 25, 2025 by ZiyueXu77 v0.47.0
2
The main issue seems to be that the main CUDA runtime library was not detected CUDA Setup likely not a BNB issue Linux low priority (will be worked on after all priority issues) Low Risk Risk of bugs in transformers and other libraries question Further information is requested
#1536 opened Feb 21, 2025 by wdh7022
@matthewdouglas
1
Ascend NPU installation Ascend NPU Related to Ascend NPU backend documentation Improvements or additions to documentation Low Risk Risk of bugs in transformers and other libraries
#1523 opened Feb 18, 2025 by LoSunny
1
Multiple issues installing for AMD GPU (Radeon RX7600XT) documentation Improvements or additions to documentation high priority (first issues that will be worked on) Low Risk Risk of bugs in transformers and other libraries ROCm
#1519 opened Feb 16, 2025 by mcondarelli
@matthewdouglas
15
torch.version.cuda.split error bug Something isn't working high priority (first issues that will be worked on) ROCm
#1513 opened Feb 13, 2025 by Lier0 v0.46.0
Default to building the C extension automatically when installing from source build contributions-welcome We welcome contributions to fix this issue! high priority (first issues that will be worked on)
#1511 opened Feb 11, 2025 by mgorny
1
@matthewdouglas
3
The output logistics of Qwen-7B under 8-bit quantization contain NaN proposing to close waiting for info
#1504 opened Feb 9, 2025 by Kairong-Han
2
Cannot load pre-quantized Janus Pro 7B bug Something isn't working medium priority (will be worked on after all high priority issues) Medium risk Risk of bugs in transformers and other libraries
#1498 opened Jan 30, 2025 by neilmehta24
ProTip! no:milestone will show everything without a milestone.