Skip to content

Pull requests: bitsandbytes-foundation/bitsandbytes

New pull request New
18 Open 672 Closed
18 Open 672 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

Fix Lion to use decoupled weight decay in the Triton 32-bit kernel Intel Optimizers Issues or feature requests relating to optimizers
#2001 opened Jul 14, 2026 by eaglstun Contributor Loading… v0.50.0
2
fix(parametrize): cache-gate hooks leak the global parametrize cache under activation checkpointing (all dequantized params retained)
#1999 opened Jul 13, 2026 by pjordanandrsn Loading…
1
fix(optim): honor user-provided max_unorm in LAMB (+ 8-bit arg guards) Optimizers Issues or feature requests relating to optimizers
#1998 opened Jul 11, 2026 by egeozkoc Contributor Loading…
Add BF16 support for int8_vectorwise_quant / LLM.int8 activation quant
#1985 opened Jul 5, 2026 by kru2710shna Loading…
1
14
Experts4bit: fused MoE weights in 4-bit NF4/FP4
#1984 opened Jul 4, 2026 by zaid646 Loading…
Fix diagnostics package version lookup
#1974 opened Jun 15, 2026 by wunianze666-netizen Loading…
Add Experts4bit for 4-bit quantization of fused MoE experts
#1965 opened Jun 5, 2026 by pjordanandrsn Loading…
1
7
fix(nn/Linear4bit): consume QuantState keys in _load_from_state_dict (#1946)
#1958 opened May 28, 2026 by Anai-Guo Loading…
1
3
docs(cuda): use Windows-native BNB_CUDA_VERSION guidance Documentation Improvements or additions to documentation Windows
#1957 opened May 27, 2026 by peter941221 Loading…
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 • Draft v0.50.0
5
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
Fix: Linear8bitLt crash with tied weights (skip_modules)
#1865 opened Feb 14, 2026 by TimDettmers Collaborator Loading…
3 tasks
1
3
fix: Stop quantize_blockwise and quantize_4bit from mutating user-provided absmax
#1863 opened Feb 14, 2026 by TimDettmers Collaborator Loading…
3 tasks done
1
1
Add k-bit blockwise quantization (K=2-5) with warp-level CUDA kernels CUDA Issues and PRs related to the CUDA backend, excluding installation/support help.
#1858 opened Feb 14, 2026 by TimDettmers Collaborator Loading…
8
Renovate pre-commit + CI a bit CI/CD
#1841 opened Jan 14, 2026 by akx Contributor • Draft
5
Enable Ascend NPU Backend with Custom Ops Integration for NF4 Support Ascend NPU Related to Ascend NPU backend
#1695 opened Jul 2, 2025 by SlightwindSec Loading…
@matthewdouglas
10
ProTip! Type g i on any issue or pull request to go back to the issue listing page.