Search results

compile error on linux with cuda 12.1

Build

CUDAIssues and PRs related to the CUDA backend, excluding installation/support help.Issues and PRs related to the CUDA backend, excluding installation/support help.

Status: Open.

#1825 In bitsandbytes-foundation/bitsandbytes;

· kforkkk opened on Dec 9, 2025

Feature Request: Support <code>.to()</code> Device Movement for Quantized Models

Waiting for Info

Status: Open.

#1815 In bitsandbytes-foundation/bitsandbytes;

· EricRollei opened on Nov 30, 2025

LARS missing in str2optimizer32bit

OptimizersIssues or feature requests relating to optimizersIssues or feature requests relating to optimizers

Status: Open.

Bug

#1810 In bitsandbytes-foundation/bitsandbytes;

· RasmusHoier opened on Nov 18, 2025

Bug Report: ROCm 8.0 binary not found (latest ROCm is 7.1)

ROCm

Waiting for Info

Status: Open.

#1799 In bitsandbytes-foundation/bitsandbytes;

· Jerry-zirui opened on Nov 5, 2025

can not import 'sync_gpu'

Status: Open.

#1792 In bitsandbytes-foundation/bitsandbytes;

· hhchang0210 opened on Oct 31, 2025

Support quantizing tensors when numel() &gt; INT_MAX

CUDAIssues and PRs related to the CUDA backend, excluding installation/support help.Issues and PRs related to the CUDA backend, excluding installation/support help.

Status: Open.

Feature

#1785 In bitsandbytes-foundation/bitsandbytes;

· matthewdouglas opened on Oct 22, 2025

·  

v0.50.0

4-bit model used more RAM than bf16 in HF transformers

Status: Open.

#1780 In bitsandbytes-foundation/bitsandbytes;

· weathon opened on Oct 7, 2025

adding unsupported NVIDIA Maxwell/Pascal/Volta architectures when using CMake &gt;=3.23.0 with CUDA13

Build

Proposing to Close

Status: Open.

#1779 In bitsandbytes-foundation/bitsandbytes;

· studioego opened on Oct 5, 2025

Reduce CUDA build matrix

Build

CUDAIssues and PRs related to the CUDA backend, excluding installation/support help.Issues and PRs related to the CUDA backend, excluding installation/support help.

Status: Open.

Task

#1778 In bitsandbytes-foundation/bitsandbytes;

· matthewdouglas opened on Oct 3, 2025

·  

v0.50.0

How to create NF4 datatype

Status: Open.

#1772 In bitsandbytes-foundation/bitsandbytes;

· phorcic opened on Oct 1, 2025

A new 8-bit quantization method (PQ-R) with 3x higher SNR for CPU

Status: Open.

#1759 In bitsandbytes-foundation/bitsandbytes;

· AlexSheff opened on Sep 22, 2025

bnb.optim.AdamW performance differs from torch.optim.AdamW despite being called with the same hyperparameters

OptimizersIssues or feature requests relating to optimizersIssues or feature requests relating to optimizers

Status: Open.

#1756 In bitsandbytes-foundation/bitsandbytes;

· inkitori opened on Sep 17, 2025