Skip to content

Pull requests: bitsandbytes-foundation/bitsandbytes

New pull request New
30 Open 703 Closed
30 Open 703 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
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[ROCm] Restore Wave64 warp size for all gfx9 targets
#2059 opened Aug 25, 2026 by 0xDELUXA Loading…
1
docs: correct Adagrad and matrix multiplication wording
#2058 opened Aug 25, 2026 by Excelius-Wang Loading…
Add Apple MPS NF4 benchmark
#2057 opened Aug 25, 2026 by hamedrabah Loading…
fix(cpu): correctly handle 1D output shape in dequantize_4bit
#2055 opened Aug 23, 2026 by SparshM8 Loading…
1
Tune SM103 fused 4-bit GEMM SIMT dispatch
#2054 opened Aug 22, 2026 by heiheiha798 Loading…
Remove redundant CUB barriers from blockwise dequantization
#2053 opened Aug 22, 2026 by heiheiha798 Loading…
Respect the current CUDA stream in blockwise quantization
#2052 opened Aug 22, 2026 by heiheiha798 Loading…
Fuse nested 4-bit scale reconstruction on SM103
#2051 opened Aug 22, 2026 by heiheiha798 Loading…
Use the current CUDA stream for non-paged optimizer updates
#2050 opened Aug 22, 2026 by heiheiha798 Loading…
Fix CPU dequantize_4bit output shape for 1-D inputs
#2048 opened Aug 20, 2026 by 2sumtech Loading…
1
ci: expand ROCm architecture coverage ROCm
#2046 opened Aug 20, 2026 by sstamenk Contributor Loading… v0.50.2
2
Fix mixed INT8 FakeTensor output metadata
#2044 opened Aug 19, 2026 by tandede Loading…
Fix decoupled weight decay ordering in the CUDA Adam/AdEMAMix kernels CUDA Issues and PRs related to the CUDA backend, excluding installation/support help. Optimizers Issues or feature requests relating to optimizers
#2040 opened Aug 14, 2026 by yentur Loading… v0.51.0
1
2
fix(autograd/MatMul4Bit): save the packed weight via save_for_backward (#2034)
#2035 opened Aug 7, 2026 by Anai-Guo Contributor Loading…
1
Warn once, and only when it matters, for unaligned gemm_4bit
#2031 opened Aug 2, 2026 by guptaishaan Loading… v0.50.2
1
2
Give each parameter its own config dict in override_config
#2029 opened Jul 29, 2026 by ErenAta16 Loading…
Run replace_linear's post_processing_function on the replacement module
#2028 opened Jul 29, 2026 by ErenAta16 Loading…
Fix int8 GEMM by using dedicated BLAS Lt handle on ROCm/CUDA ROCm
#2018 opened Jul 23, 2026 by zjin-lcf Contributor Loading…
2 of 3 tasks
1
Add BF16 support for int8_vectorwise_quant / LLM.int8 activation quant
#1985 opened Jul 5, 2026 by kru2710shna Loading…
1
14
Add Experts4bit for 4-bit quantization of fused MoE experts
#1965 opened Jun 5, 2026 by pjordanandrsn Contributor Loading…
1
9
fix(nn/Linear4bit): consume QuantState keys in _load_from_state_dict (#1946)
#1958 opened May 28, 2026 by Anai-Guo Contributor Loading…
1
3
refactor: Rewrite assert statements as exceptions
#1931 opened Apr 23, 2026 by rapsealk Contributor Loading…
5 of 7 tasks
1
ci: add nightly integration tests for transformers, accelerate, peft CI/CD
#1923 opened Apr 13, 2026 by Titus-von-Koeller Collaborator Loading…
1
[ROCm] Rework ROCm build to use ROCm version instead of HIP version ROCm
#1888 opened Mar 4, 2026 by sstamenk Contributor Loading… v0.50.2
6
docs: add an energy-efficiency FAQ entry for quantization Documentation Improvements or additions to documentation
#1882 opened Feb 24, 2026 by hongping-zh Loading…
10
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.