Skip to content

Pull requests: auroralabs-loci/llama.cpp

New pull request New
933 Open 154 Closed
933 Open 154 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

UPSTREAM PR #18755: Kimi-Linear support (backend agnostic + MLA KV cache)
#1087 opened Jan 31, 2026 by loci-dev Loading…
UPSTREAM PR #19141: [WIP]ggml-hexagon: flash-attn opt - part2
#1086 opened Jan 31, 2026 by loci-dev Loading…
UPSTREAM PR #19167: model : support LongCat-Flash-Lite (ngram embeddings)
#1085 opened Jan 31, 2026 by loci-dev Loading…
UPSTREAM PR #19209: ggml-cpu: split across kv for faster TG
#1084 opened Jan 31, 2026 by loci-dev Loading…
UPSTREAM PR #19218: nix: fix nix develop .#python-scripts
#1082 opened Jan 30, 2026 by loci-dev Loading…
1
UPSTREAM PR #19209: ggml-cpu: split across kv for faster TG
#1081 opened Jan 30, 2026 by loci-dev Loading…
1
UPSTREAM PR #19205: Add missing unordered_map include to jinja/value.h
#1080 opened Jan 30, 2026 by loci-dev Loading…
1
UPSTREAM PR #19206: metal: convert shutdown assertion to warning log
#1079 opened Jan 30, 2026 by loci-dev Loading…
1
UPSTREAM PR #19199: memory : remove unused tmp_buf
#1078 opened Jan 30, 2026 by loci-dev Loading…
1
UPSTREAM PR #19196: ggml-cpu: optimize q4_0_q8_0 scales using Zvfhmin
#1077 opened Jan 30, 2026 by loci-dev Loading…
6
UPSTREAM PR #19195: Remove pipeline cache mutexes
#1076 opened Jan 30, 2026 by loci-dev Loading…
2
UPSTREAM PR #19194: Correctly fetch q8_1 quantize pipeline in test as needed by 8a3519b
#1075 opened Jan 30, 2026 by loci-dev Loading…
1
UPSTREAM PR #19183: [RFC] implementing Expected Attention in llama.cpp?
#1074 opened Jan 29, 2026 by loci-dev Loading…
2
UPSTREAM PR #19186: add tensor type checking as part of cuda graph properties
#1073 opened Jan 29, 2026 by loci-dev Loading…
1
UPSTREAM PR #19182: model: support Longcat-Flash (help wanted)
#1072 opened Jan 29, 2026 by loci-dev Loading…
1
UPSTREAM PR #19178: Finish “TODO: avoid using atexit() here by making console a singleton”
#1071 opened Jan 29, 2026 by loci-dev Loading…
UPSTREAM PR #19176: jinja : do not pass empty tools and add some none filters
#1070 opened Jan 29, 2026 by loci-dev Loading…
UPSTREAM PR #19132: ggml: aarch64: Implement SVE in Gemm q4_k 8x8 q8_k Kernel
#1069 opened Jan 29, 2026 by loci-dev Loading…
3
UPSTREAM PR #19170: Add Kimi-K2.5 support
#1068 opened Jan 29, 2026 by loci-dev Loading…
1
UPSTREAM PR #19168: build(cuda): Add warning for CUDA 13.0 Blackwell compiler bug
#1067 opened Jan 29, 2026 by loci-dev Loading…
1
UPSTREAM PR #19167: model : support LongCat-Flash-Lite (ngram embeddings)
#1066 opened Jan 28, 2026 by loci-dev Loading…
1
UPSTREAM PR #19165: cuda : fix nkvo
#1065 opened Jan 28, 2026 by loci-dev Loading…
1
UPSTREAM PR #19166: Update server-models.cpp
#1064 opened Jan 28, 2026 by loci-dev Loading…
1
UPSTREAM PR #19164: spec : add ngram-mod
#1063 opened Jan 28, 2026 by loci-dev Loading…
1 task
2
UPSTREAM PR #19162: Fix typos in SYCL documentation
#1062 opened Jan 28, 2026 by loci-dev Loading…
1
ProTip! Filter pull requests by the default branch with base:overlay.