Skip to content

Pull requests: TheTom/llama-cpp-turboquant

New pull request New
6 Open 35 Closed
6 Open 35 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

vulkan: TQ4_1s support for model weights ggml testing Vulkan
#69 opened Apr 11, 2026 by Titaniumtown Loading…
2
fix: inverse WHT in test-turbo-quant.c round-trip (#59) ggml testing
#63 opened Apr 9, 2026 by seanrasch Loading…
perf: turbo VEC flash attention — +9% decode on CUDA via autoresearch ggml Nvidia GPU script
#53 opened Apr 4, 2026 by signalnine Loading…
7 tasks done
6
metal: MUL_MAT kernels for turbo3/turbo4 + dual-LUT dequant (port of PR #22) Apple Metal ggml
#49 opened Apr 4, 2026 by TheTom Owner Loading…
1
fix: HIP/ROCm compatibility — check cudaMemcpyToSymbol errors, guard … ggml Nvidia GPU
#41 opened Apr 1, 2026 by terrysimons • Draft
1
feat: ROCm/HIP support for turbo3 KV cache (gfx1100/RDNA3) documentation Improvements or additions to documentation ggml Nvidia GPU
#5 opened Mar 27, 2026 by apollosenvy Loading…
5 of 7 tasks
15
ProTip! Adding no:label will show everything without a label.