| 2026-06-07 |
Sigbjørn Skjæret | convert : fix Gemma4 with no audio encoder (#24242) |
commit | commitdiff | tree |
| 2026-06-07 |
Sigbjørn Skjæret | docker : bump cuda13 to 13.3.0 (#24228) |
commit | commitdiff | tree |
| 2026-06-06 |
Tarek Dakhran | common/chat : fix LFM2/LFM2.5 reasoning round-trip... |
commit | commitdiff | tree |
| 2026-06-06 |
Xuan-Son Nguyen | mtmd: support "frame merge" for qwen-vl-based models... |
commit | commitdiff | tree |
| 2026-06-06 |
Adrien Gallouët | completion : remove useless statics (#24226) |
commit | commitdiff | tree |
| 2026-06-06 |
Adrien Gallouët | completion : fix format specifier in LOG_INF (#24213) |
commit | commitdiff | tree |
| 2026-06-06 |
Xuan-Son Nguyen | mtmd, server: add "placeholder bitmap" for counting... |
commit | commitdiff | tree |
| 2026-06-06 |
Ruben Ortlam | vulkan: check coopmat2 features before reporting suppor... |
commit | commitdiff | tree |
| 2026-06-06 |
Sigbjørn Skjæret | model : rename local n_layer_all variable (#24209) |
commit | commitdiff | tree |
| 2026-06-06 |
Sigbjørn Skjæret | context : fix off-by-one comparisons to n_gpu_layers... |
commit | commitdiff | tree |
| 2026-06-05 |
lhez | opencl: improve get_rows, cpy, concat and q6_k flat... |
commit | commitdiff | tree |
| 2026-06-05 |
Tarek Dakhran | common/chat : unify and fix LFM2/LFM2.5 tool parser... |
commit | commitdiff | tree |
| 2026-06-05 |
Ruben Ortlam | vulkan: add fwht support for Intel with shmem reduction... |
commit | commitdiff | tree |
| 2026-06-05 |
Xuan-Son Nguyen | model: fix build failed (#24193) |
commit | commitdiff | tree |
| 2026-06-05 |
Gabe Goodhart | model, mtmd: Granite4 Vision (#23545) |
commit | commitdiff | tree |
| 2026-06-05 |
Johannes Gäßler | TP: round up granularity to 128 (#24180) |
commit | commitdiff | tree |
| 2026-06-05 |
therealkenc | cli: fix model params not propagated (#23893) |
commit | commitdiff | tree |
| 2026-06-05 |
Georgi Gerganov | model : fix llama_model::n_gpu_layers() (#24188) |
commit | commitdiff | tree |
| 2026-06-05 |
Pascal | ui: run npm install when package-lock.json is newer... |
commit | commitdiff | tree |
| 2026-06-05 |
Mario | Fix link to available UI settings (#24169) |
commit | commitdiff | tree |
| 2026-06-05 |
Xuan-Son Nguyen | ui: add ignore-scripts=true to npmrc (#24149) |
commit | commitdiff | tree |
| 2026-06-05 |
Pedro Cuenca | docs: Update quantization readme (#24133) |
commit | commitdiff | tree |
| 2026-06-05 |
Georgi Gerganov | minor : fix lint issues (#24165) |
commit | commitdiff | tree |
| 2026-06-05 |
Georgi Gerganov | hparams : refactor `hparams.n_layer` (#24060) |
commit | commitdiff | tree |
| 2026-06-05 |
Charles Xu | kleidiai : dynamic chunck-based scheduling for hybrid... |
commit | commitdiff | tree |
| 2026-06-05 |
Oliver Simons | CUDA: enroll mul_mat_vec_q_moe into pdl (#24087) |
commit | commitdiff | tree |
| 2026-06-05 |
Daniel Bevenius | ci : build-msys job slimming [no ci] (#24157) |
commit | commitdiff | tree |
| 2026-06-05 |
Mason Milburn | sycl : port multi-column MMVQ from CUDA backend (#21845) |
commit | commitdiff | tree |
| 2026-06-04 |
Georgi Gerganov | server : disable on-device spec checkpoints (#24108) |
commit | commitdiff | tree |
| 2026-06-04 |
Xuan-Son Nguyen | arg: fix double mtp downloads (#24128) |
commit | commitdiff | tree |
| 2026-06-04 |
viggy | webui: [a11y] fix keyboard navigation issues in chat... |
commit | commitdiff | tree |
| 2026-06-04 |
Bartowski | Move duplicated imatrix code into single common imatrix... |
commit | commitdiff | tree |
| 2026-06-04 |
Aleksander... | ui: Fixed packages (#24119) |
commit | commitdiff | tree |
| 2026-06-04 |
MagicExists | ui: added single line reasoning preview (#23601) |
commit | commitdiff | tree |
| 2026-06-04 |
forforever73 | return filter to save memory (#24125) |
commit | commitdiff | tree |
| 2026-06-04 |
Pedro Cuenca | convert: Fix Gemma 4 Unified conversion (#24118) |
commit | commitdiff | tree |
| 2026-06-04 |
Kartik Sirohi | ggml: vectorize ggml_vec_dot_q4_1_q8_1 with WASM SIMD12... |
commit | commitdiff | tree |
| 2026-06-04 |
Yongyue Sun | server: avoid unnecessary checkpoint restore when new... |
commit | commitdiff | tree |
| 2026-06-04 |
Xuan-Son Nguyen | agents: refactor, include more guidelines (#24111) |
commit | commitdiff | tree |
| 2026-06-04 |
Pascal | webui: fix tool selector toggle/counter, key tools... |
commit | commitdiff | tree |
| 2026-06-04 |
Gerard Martinez | build : use umbrella Headers directory for XCFramework... |
commit | commitdiff | tree |
| 2026-06-04 |
A B | server : add header to tools/server/server-http.h ... |
commit | commitdiff | tree |
| 2026-06-04 |
Andrea Richiardi | cmake: skip cvector-generator and export-lora when... |
commit | commitdiff | tree |
| 2026-06-04 |
Andrei | fix(mtmd): handle Gemma 4 audio projector embedding... |
commit | commitdiff | tree |
| 2026-06-04 |
Georgi Gerganov | readme : add status badges (#24104) |
commit | commitdiff | tree |
| 2026-06-04 |
Georgi Gerganov | tests : refactor test-save-load-state to accept token... |
commit | commitdiff | tree |
| 2026-06-04 |
Georgi Gerganov | metal : reduce rset heartbeat from 500ms -> 5ms (#24074) |
commit | commitdiff | tree |
| 2026-06-04 |
Reese Levine | ggml-webgpu: FlashAttention refactor + standardize... |
commit | commitdiff | tree |
| 2026-06-04 |
rehan-10xengineer | ggml-cpu: extend RVV quantization vec dot to higher... |
commit | commitdiff | tree |
| 2026-06-04 |
Todd Malsbary | sycl : Improve SYCL doc (#23025) |
commit | commitdiff | tree |
| 2026-06-03 |
Andrei | mtmd: fix Gemma 4 unified FPE (#24088) |
commit | commitdiff | tree |
| 2026-06-03 |
Aman Gupta | qwen35: use post-norm hidden state for MTP (#24025) |
commit | commitdiff | tree |
| 2026-06-03 |
Xuan-Son Nguyen | mtmd: enable non-causal vision for gemma 4 unified... |
commit | commitdiff | tree |
| 2026-06-03 |
Xuan-Son Nguyen | mtmd, model: allow skip build_vit() (#24077) |
commit | commitdiff | tree |
| 2026-06-03 |
Aleksander... | ui: Mermaid Diagrams in chat + interactive preview... |
commit | commitdiff | tree |
| 2026-06-03 |
Andreas Kieslinger | Avoid PDL race conditions by disabling __restrict__... |
commit | commitdiff | tree |
| 2026-06-03 |
Charles Xu | ggml-cpu: use runtime SVE width in FWHT (#24059) |
commit | commitdiff | tree |
| 2026-06-03 |
Aman Gupta | cuda: reserve space for quantize kv-cache at startup... |
commit | commitdiff | tree |
| 2026-06-03 |
Georgi Gerganov | tests : add support for qwen3 SSM archs (#24031) |
commit | commitdiff | tree |
| 2026-06-03 |
Alessandro... | update BoringSSL to 0.20260526.0 (#23794) |
commit | commitdiff | tree |
| 2026-06-03 |
Georgi Gerganov | ci : disable ccache for msvc windows release jobs ... |
commit | commitdiff | tree |
| 2026-06-03 |
Ryan Mangeno | arg : removed unecesary mmproj download when users... |
commit | commitdiff | tree |
| 2026-06-02 |
lhez | opencl: use flat variants of q4_K and q6_K gemv for... |
commit | commitdiff | tree |
| 2026-06-02 |
Max Krasnyansky | hexagon: profiler output fix and script updates (#24042) |
commit | commitdiff | tree |
| 2026-06-02 |
Mikhail Podvitskii | model: add Mellum architecture (#23966) |
commit | commitdiff | tree |
| 2026-06-02 |
Hans Florian | model : support granite multilingual embeddings R2... |
commit | commitdiff | tree |
| 2026-06-02 |
Piotr Wilkin... | StepFun 3.5 MTP (#23274) |
commit | commitdiff | tree |
| 2026-06-02 |
Daniel Bevenius | common : fix state save in common_prompt_batch_decode... |
commit | commitdiff | tree |
| 2026-06-02 |
Xuan-Son Nguyen | server: add SSE ping interval (#24013) |
commit | commitdiff | tree |
| 2026-06-02 |
Georgi Gerganov | ci : reduce self-hosted server workflow jobs (#24012) |
commit | commitdiff | tree |
| 2026-06-02 |
Mikhail Podvitskii | docs : update HOWTO-add-model.md (#23883) |
commit | commitdiff | tree |
| 2026-06-02 |
Marcos Del... | ui: simplify network error handling (#23431) |
commit | commitdiff | tree |
| 2026-06-02 |
Aleksander... | ui: Add Thinking mode toggle with reasoning effort... |
commit | commitdiff | tree |
| 2026-06-02 |
Georgi Gerganov | kv-cache : SWA checkpoints store only non-masked cells... |
commit | commitdiff | tree |
| 2026-06-02 |
forforever73 | convert : support Step3.7-Flash (#23845) |
commit | commitdiff | tree |
| 2026-06-02 |
Georgi Gerganov | llama : deprecate `llama_set_warmup` (#24009) |
commit | commitdiff | tree |
| 2026-06-02 |
Max Krasnyansky | hexagon: MUL_MAT, MUL_MAT_ID, FLASH_ATTN and GDN cleanu... |
commit | commitdiff | tree |
| 2026-06-02 |
Todor Boinovski | hexagon: add gelu_quick (#24007) |
commit | commitdiff | tree |
| 2026-06-02 |
Pascal | server: real-time reasoning interruption via control... |
commit | commitdiff | tree |
| 2026-06-02 |
Anav Prasad | clean up unused variables warnings (#23975) |
commit | commitdiff | tree |
| 2026-06-02 |
lhez | opencl: fix compiler warnings for non-adreno path ... |
commit | commitdiff | tree |
| 2026-06-01 |
Masashi Yoshimura | revert to using global_invocation_id for cpy shader... |
commit | commitdiff | tree |
| 2026-06-01 |
Georgi Gerganov | speculative : fix n_outputs_max and remove draft-simple... |
commit | commitdiff | tree |
| 2026-06-01 |
Christian Hoener... | nix : add nix-nodejs facilities to build Web UI (#23846) |
commit | commitdiff | tree |
| 2026-06-01 |
shaofeiqi | opencl: add basic support for q5_0 and q5_1 (#23548) |
commit | commitdiff | tree |
| 2026-06-01 |
Adrien Gallouët | vendor : update cpp-httplib to 0.46.1 (#23980) |
commit | commitdiff | tree |
| 2026-06-01 |
Aman Gupta | llama: limit max outputs of `llama_context` (#23861) |
commit | commitdiff | tree |
| 2026-06-01 |
Shrivas Shankar | metal: template GLU kernels to support f16/f32 (#23882) |
commit | commitdiff | tree |
| 2026-06-01 |
Jeff Bolz | vulkan: don't hold the device mutex while compiling... |
commit | commitdiff | tree |
| 2026-06-01 |
Winston Ma | vulkan: reduce host memory lock contention (#23376) |
commit | commitdiff | tree |
| 2026-06-01 |
o7si | vocab: add normalizer.lowercase support to WPM (#23899) |
commit | commitdiff | tree |
| 2026-06-01 |
Johannes Gäßler | TP: quantized KV cache support (#23792) |
commit | commitdiff | tree |
| 2026-06-01 |
Georgi Gerganov | security : disable private disclosures (#23963) |
commit | commitdiff | tree |
| 2026-06-01 |
Junwon Hwang | model: Add EXAONE 4.5 implementations (#21733) |
commit | commitdiff | tree |
| 2026-06-01 |
Matt Corallo | vulkan: Block-load Q3_K/Q6_K block data and subtract... |
commit | commitdiff | tree |
| 2026-06-01 |
Winston Ma | vulkan: Removed unused functions (#23175) |
commit | commitdiff | tree |
| 2026-06-01 |
Aldehir Rojas | common : support manually triggering the reasoning... |
commit | commitdiff | tree |
| 2026-06-01 |
Georgi Gerganov | ci : add missing Linux label to cpu-x64-high-perf runne... |
commit | commitdiff | tree |
| 2026-06-01 |
Neo Zhang | [SYCL] Support Q4_1, Q5_0, Q5_1 in Flash-attention... |
commit | commitdiff | tree |
| 2026-06-01 |
Neo Zhang | [SYCL] Add more types in GET_ROWS OP (#23710) |
commit | commitdiff | tree |
| next |
overview / pkg / ggml / sources / llama.cpp / shortlog