Skip to content

Pull requests: ggerganov/llama.cpp

New pull request New
321 Open 4,322 Closed
321 Open 4,322 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Opt class for positional argument handling examples
#10508 opened Nov 26, 2024 by ericcurtin Loading…
2 of 4 tasks
vulkan: skip integer div/mod in get_offsets for batch_idx==0
#10506 opened Nov 26, 2024 by jeffbolznv Loading…
2 of 4 tasks
speculative : simplify the implementation examples
#10504 opened Nov 25, 2024 by ggerganov • Draft
vulkan: get the first command buffer submitted sooner
#10499 opened Nov 25, 2024 by jeffbolznv Loading…
2 of 4 tasks
1
Add some minimal optimizations for CDNA Nvidia GPU Issues specific to Nvidia GPUs
#10498 opened Nov 25, 2024 by IMbackK Loading…
1 of 3 tasks
6
vulkan: fix group_norm testing Everything test related
#10496 opened Nov 25, 2024 by jeffbolznv Loading…
2 of 4 tasks
1
imatrix-combine-only idea examples
#10492 opened Nov 25, 2024 by robbiemu Loading…
[CANN] RoPE optimization Ascend NPU issues specific to Ascend NPUs
#10488 opened Nov 25, 2024 by noemotiovon Loading…
2 of 4 tasks
ggml-cpu: cmake add arm64 cpu feature check for macos
#10487 opened Nov 25, 2024 by chaxu01 Loading…
2
vulkan: further optimize q5_k mul_mat_vec
#10479 opened Nov 25, 2024 by jeffbolznv Loading…
2 of 4 tasks
2
cmake : enable warnings in llama build Compilation issues examples Nvidia GPU Issues specific to Nvidia GPUs
#10474 opened Nov 24, 2024 by ggerganov Loading…
6
vulkan: Handle GPUs with less shared memory
#10468 opened Nov 23, 2024 by jeffbolznv Loading…
2 of 4 tasks
1
vulkan: optimize Q2_K and Q3_K mul_mat_vec
#10459 opened Nov 23, 2024 by jeffbolznv Loading…
2 of 4 tasks
ci : add ubuntu cuda build, build with one arch on windows devops improvements to build systems and github actions nix Issues specific to consuming flake.nix, or generally concerned with ❄ Nix-based llama.cpp deployment
#10456 opened Nov 22, 2024 by slaren Loading…
[CANN] Improve the Inferencing Performance for Ascend NPU Device Ascend NPU issues specific to Ascend NPUs
#10454 opened Nov 22, 2024 by shen-shanshan Loading…
2 of 4 tasks
10
llava: return false instead of exit examples
#10452 opened Nov 22, 2024 by tinglou Loading…
2 of 4 tasks
1
fix: ggml: fix vulkan-shaders-gen build
#10448 opened Nov 22, 2024 by sparkleholic Loading…
2 of 4 tasks
4
Refactor/online repacking ggml changes relating to the ggml tensor library for machine learning
#10446 opened Nov 21, 2024 by Djip007 • Draft
2 of 4 tasks
19
Integrating llama.cpp with Microsoft Word
#10443 opened Nov 21, 2024 by GPTLocalhost Loading…
2 of 4 tasks
1
vulkan: define all quant data structures in types.comp
#10440 opened Nov 20, 2024 by jeffbolznv Loading…
2 of 4 tasks
server : (proposal) allow user to customize chat format examples server
#10438 opened Nov 20, 2024 by ngxson • Draft
2 tasks done
1
sycl : offload of get_rows set to false SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language
#10432 opened Nov 20, 2024 by Alcpz Loading…
2 of 4 tasks
4
bug-fix: snprintf prints NULL in place of the last character examples server
#10419 opened Nov 20, 2024 by kallewoof Loading…
2 of 4 tasks
9
server : replace behave with pytest devops improvements to build systems and github actions examples nix Issues specific to consuming flake.nix, or generally concerned with ❄ Nix-based llama.cpp deployment python python script changes server
#10416 opened Nov 19, 2024 by ngxson Loading…
5 tasks done
4
sycl : permuted mul_mat through oneMKL SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language
#10408 opened Nov 19, 2024 by Alcpz Loading…
2 of 4 tasks
1
ProTip! Add no:assignee to see everything that’s not assigned.