-
-
Notifications
You must be signed in to change notification settings - Fork 5.9k
Pull requests: unslothai/unsloth
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Studio: report the real llama-server context window and add an opt-in overflow policy for OpenAI-compatible serving
#6164
opened Jun 10, 2026 by
danielhanchen
Member
Loading…
fix: avoid false flash-attn warning with transformers 5
#6161
opened Jun 10, 2026 by
pirrozani
Loading…
Add FastDiffusionModel slow path for text-diffusion models (DiffusionGemma)
#6158
opened Jun 10, 2026 by
danielhanchen
Member
Loading…
Keep audio feature extractors right padded when loading processors
#6157
opened Jun 10, 2026 by
danielhanchen
Member
Loading…
Expose runtime context length for hub models
#6154
opened Jun 10, 2026 by
alkinun
Contributor
Loading…
fix: avoid blocking the event loop with time.sleep in async load_checkpoint
#6135
opened Jun 10, 2026 by
kratos0718
Loading…
fix(studio): fall back to copy when os.replace is blocked during install activation
#6133
opened Jun 10, 2026 by
Anai-Guo
Contributor
Loading…
Fix bitsandbytes ROCm GPU arch and warp size detection on Windows
#6127
opened Jun 10, 2026 by
danielhanchen
Member
Loading…
fix(rocm): cap RDNA2 to rocm6.4 torch and fix inference crash on gfx1030-gfx1036
#6124
opened Jun 9, 2026 by
LeoBorcherding
Contributor
•
Draft
fix(rocm): stop overwriting ROCR_VISIBLE_DEVICES in apply_gpu_ids
#6123
opened Jun 9, 2026 by
LeoBorcherding
Contributor
•
Draft
Studio: Add custom provider option to Connections
#6112
opened Jun 9, 2026 by
Imagineer99
Collaborator
Loading…
Studio: improve logging for dynamic transformers version switching
#6108
opened Jun 9, 2026 by
danielhanchen
Member
Loading…
feat(export): GGUF shard size control, save directory override, fix private Hub push
#6107
opened Jun 8, 2026 by
LeoBorcherding
Contributor
•
Draft
6 tasks
feat: Intel Arc / XPU GPU support for Studio and Gemma training
#6106
opened Jun 8, 2026 by
LeoBorcherding
Contributor
•
Draft
Studio: ignore unsupported env proxy during Studio startup
#6102
opened Jun 8, 2026 by
Imagineer99
Collaborator
•
Draft
Studio: tune llama.cpp env for data-center GPUs
#6098
opened Jun 8, 2026 by
danielhanchen
Member
Loading…
patch: fix EmptyLogits gathering in nested payloads and Accelerate recursively_apply
#6092
opened Jun 8, 2026 by
MdHussain121
Loading…
fix: respect absolute export paths to prevent cross-drive copy failures (WinError 112)
#6088
opened Jun 8, 2026 by
anmolxlight
Contributor
Loading…
Fix/ Allow empty tool content and system role in API messages
#6052
opened Jun 6, 2026 by
BardiaKoopah
Loading…
fix(studio/responses): emit placeholder for empty tool output to avoid 500
#6049
opened Jun 6, 2026 by
Anai-Guo
Contributor
Loading…
2 tasks
Studio: Add Tensor-Parallel llama.cpp support
#6040
opened Jun 6, 2026 by
oobabooga
Collaborator
Loading…
fix(setup): recognize Visual Studio Build Tools 2026 (v18) in Find-VsBuildTools
#6038
opened Jun 5, 2026 by
LeoBorcherding
Contributor
•
Draft
3 tasks
Previous Next
ProTip!
Filter pull requests by the default branch with base:main.