| 2026-05-18 |
Pranav Dhinakar | ggml-hexagon: add PAD op HVX kernel (#23078) |
commit | commitdiff | tree |
| 2026-05-18 |
SamareshSingh | docker : add OCI image labels for version and build... |
commit | commitdiff | tree |
| 2026-05-18 |
Adrien Gallouët | common : remove hf cache migration (#23266) |
commit | commitdiff | tree |
| 2026-05-18 |
Aleksander... | ui: Update KaTeX package and clean up logs from `sass... |
commit | commitdiff | tree |
| 2026-05-18 |
Aleksander... | feat: add scroll-to-bottom button to chat + prevent... |
commit | commitdiff | tree |
| 2026-05-18 |
Aleksander... | ui: Refactor models store, MCP service, and gate logs... |
commit | commitdiff | tree |
| 2026-05-18 |
Aleksander... | ui: Centralize monospace font styles in app.css (#23272) |
commit | commitdiff | tree |
| 2026-05-18 |
Martin Andersson | webui: fix Tailwind v4 utility classes missing when... |
commit | commitdiff | tree |
| 2026-05-18 |
Andrei | llama: initialize pre-norm embedding mask flag (#23256) |
commit | commitdiff | tree |
| 2026-05-18 |
Sigbjørn Skjæret | add myself to conversion (#23261) |
commit | commitdiff | tree |
| 2026-05-18 |
Martin Klacer | ci : added kleidiai-server to server-self-hosted workfl... |
commit | commitdiff | tree |
| 2026-05-18 |
Georgi Gerganov | scripts : allow wc2wt with an existing branch (#23189) |
commit | commitdiff | tree |
| 2026-05-18 |
Intel AI Get... | sycl: scalar SWAR byte-subtract in Q6_K MMVQ dot produc... |
commit | commitdiff | tree |
| 2026-05-18 |
Intel AI Get... | sycl: route small f32 matmuls to oneMKL, bypass oneDNN... |
commit | commitdiff | tree |
| 2026-05-18 |
Neo Zhang | sycl : fix error when use -mg 1 error (#23140) |
commit | commitdiff | tree |
| 2026-05-18 |
Incarnas | update bid to match each layers MTP source (#23237) |
commit | commitdiff | tree |
| 2026-05-18 |
Sigbjørn Skjæret | cmake : do not check for bin install dir (#23234) |
commit | commitdiff | tree |
| 2026-05-17 |
Gabe Goodhart | feat: Support d_conv=15 for ssm-conv.cu (#23017) |
commit | commitdiff | tree |
| 2026-05-17 |
Aldehir Rojas | cmake : fix LLAMA_BUILD_UI logic (#23190) |
commit | commitdiff | tree |
| 2026-05-17 |
Sigbjørn Skjæret | cmake : do not install conversion script (#23204) |
commit | commitdiff | tree |
| 2026-05-17 |
Oliver Simons | CUDA: Continue directly including cuda/iterator (#23102) |
commit | commitdiff | tree |
| 2026-05-17 |
Aman Gupta | llama: avoid copying logits during prompt decode in... |
commit | commitdiff | tree |
| 2026-05-17 |
Aldehir Rojas | common : delegate assistant continuation to underlying... |
commit | commitdiff | tree |
| 2026-05-17 |
Jan Ekström | ggml-vulkan/CMakeLists: add a check for SPIRV-Headers... |
commit | commitdiff | tree |
| 2026-05-17 |
Pascal | vulkan: add cpy bf16 -> f32 pipelines (#22677) |
commit | commitdiff | tree |
| 2026-05-17 |
Jeff Bolz | vulkan: Support unaligned tensors for ROPE (#22637) |
commit | commitdiff | tree |
| 2026-05-17 |
Aldehir Rojas | common : enable streaming JSON argument values (#23173) |
commit | commitdiff | tree |
| 2026-05-17 |
Jeff Bolz | vulkan: fuse SSM_CONV + BIAS + SILU (#22653) |
commit | commitdiff | tree |
| 2026-05-17 |
Rares Vernica | server : honor --embd-normalize CLI arg (#23125) |
commit | commitdiff | tree |
| 2026-05-17 |
ddh0 | ngram : reduce noisy logs (#23185) |
commit | commitdiff | tree |
| 2026-05-17 |
Judd | webui: support video files as input (#22830) |
commit | commitdiff | tree |
| 2026-05-16 |
Xuan-Son Nguyen | server: (router) alloc tmp buffer on heap (#23159) |
commit | commitdiff | tree |
| 2026-05-16 |
Pascal | server: skip device enumeration in router mode to avoid... |
commit | commitdiff | tree |
| 2026-05-16 |
Winston Ma | vulkan: removed duplicate #include <memory> in headers... |
commit | commitdiff | tree |
| 2026-05-16 |
Aleksander... | ui: Add request timeout for MCP tool calls (#23138) |
commit | commitdiff | tree |
| 2026-05-16 |
Georgi Gerganov | sync : ggml |
commit | commitdiff | tree |
| 2026-05-16 |
Georgi Gerganov | ggml : bump version to 0.12.0 (ggml/1494) |
commit | commitdiff | tree |
| 2026-05-16 |
CrispStrobe | metal : tighten input-position loop in kernel_conv_tran... |
commit | commitdiff | tree |
| 2026-05-16 |
Steve Lhomme | ggml: install ggml.pc in <libdir>/pkgconfig (ggml/1480) |
commit | commitdiff | tree |
| 2026-05-16 |
Holger Voormann | ui: Correct links in `tools/ui/README.md` [no ci] ... |
commit | commitdiff | tree |
| 2026-05-16 |
Alessandro... | vendor : update cpp-httplib to 0.45.0 (#23103) |
commit | commitdiff | tree |
| 2026-05-16 |
Aman Gupta | llama + spec: MTP Support (#22673) |
commit | commitdiff | tree |
| 2026-05-16 |
kubawoo | ui: Fix handling of MCP resource template parameters... |
commit | commitdiff | tree |
| 2026-05-16 |
viggy | webui : [ChatFormActionAdd][a11y] fix accessibility... |
commit | commitdiff | tree |
| 2026-05-16 |
Pascal | ui: untrack settings sync in props effect to prevent... |
commit | commitdiff | tree |
| 2026-05-16 |
Aleksander... | fix: Add build step using build workflow to publish... |
commit | commitdiff | tree |
| 2026-05-16 |
ynankani | model : NvFP4 quantized LM head support (#23046) |
commit | commitdiff | tree |
| 2026-05-16 |
Aleksander... | ui: Restructure repo to use `tools/ui` folder and ... |
commit | commitdiff | tree |
| 2026-05-15 |
Sigbjørn Skjæret | ci : fix release symlinks (#23119) |
commit | commitdiff | tree |
| 2026-05-15 |
Omer Ozarslan | webui: Use lowercase hash for HF checksum check (#23107) |
commit | commitdiff | tree |
| 2026-05-15 |
Pascal | tests: add BF16 non-contig coverage for MUL_MAT permuta... |
commit | commitdiff | tree |
| 2026-05-15 |
Julien Chaumond | docs: document `usage` object in server timings respons... |
commit | commitdiff | tree |
| 2026-05-15 |
Xuan-Son Nguyen | mtmd: add chunks and fix preproc for qwen3a (#23073) |
commit | commitdiff | tree |
| 2026-05-15 |
Pascal | webui: fix theme from --webui-config-file not applied... |
commit | commitdiff | tree |
| 2026-05-15 |
Sigbjørn Skjæret | convert : fix Qwen3 ASR conversion (#23081) |
commit | commitdiff | tree |
| 2026-05-15 |
Piotr Wilkin... | Refactor: convert_hf_to_gguf.py (#17114) |
commit | commitdiff | tree |
| 2026-05-15 |
Sigbjørn Skjæret | ci : fix transform of top . entry in release archive... |
commit | commitdiff | tree |
| 2026-05-15 |
Georgi Gerganov | llama-eval : add AIME 2026 dataset support (#23058) |
commit | commitdiff | tree |
| 2026-05-15 |
Aman Gupta | reasoning-budget: clone should do a deep-copy (#23095) |
commit | commitdiff | tree |
| 2026-05-15 |
Pascal | webui: preserve partial response on streaming error... |
commit | commitdiff | tree |
| 2026-05-15 |
Sid Shaytay | Support for Codex CLI by skipping unsupported Responses... |
commit | commitdiff | tree |
| 2026-05-15 |
KITAITI Makoto | readme : update bindings (#23063) |
commit | commitdiff | tree |
| 2026-05-14 |
Pranav Dhinakar | ggml-hexagon: cpy: add contiguous fast-path in reshape... |
commit | commitdiff | tree |
| 2026-05-14 |
Johannes Gäßler | HIP: RDNA3 mma FA, faster AMD transpose, tune AMD ... |
commit | commitdiff | tree |
| 2026-05-14 |
Zack Li | CI : support IOT device (IQ9) (#22987) |
commit | commitdiff | tree |
| 2026-05-14 |
Reese Levine | ggml-webgpu: Enable NVIDIA self-hosted CI (#22976) |
commit | commitdiff | tree |
| 2026-05-14 |
Zheyuan Chen | ggml-webgpu: makes the flash attn vec path subgroup... |
commit | commitdiff | tree |
| 2026-05-14 |
Aleksander... | fix: Propagate version tag to WebUI asset download... |
commit | commitdiff | tree |
| 2026-05-14 |
Aman Gupta | contributing: new contributors should not submit trivia... |
commit | commitdiff | tree |
| 2026-05-14 |
Aleksander... | webui: Move static build output from repo code to HF... |
commit | commitdiff | tree |
| 2026-05-14 |
Georgi Gerganov | logs : reduce (#23021) |
commit | commitdiff | tree |
| 2026-05-14 |
alex-spacemit | ggml-cpu: Add IME2 Instruction Support for the SpacemiT... |
commit | commitdiff | tree |
| 2026-05-14 |
Neo Zhang | docker : revert stable version of intel compute-runtime... |
commit | commitdiff | tree |
| 2026-05-14 |
Kabir Potdar | unicode,test: add Qwen3.5 non-backtracking tokenizer... |
commit | commitdiff | tree |
| 2026-05-14 |
Ruben Ortlam | vulkan: fix matmul integer pipeline selection (#23005) |
commit | commitdiff | tree |
| 2026-05-14 |
Aleksander... | fix: Autoscroll detection (#23026) |
commit | commitdiff | tree |
| 2026-05-14 |
Katostrofik | SYCL: fix multi-GPU system RAM exhaustion by using... |
commit | commitdiff | tree |
| 2026-05-13 |
Zheyuan Chen | ggml-webgpu: only use subgroup-matrix path when head... |
commit | commitdiff | tree |
| 2026-05-13 |
scutler-nv | Fix for issue #22974. Cast intermediate results to... |
commit | commitdiff | tree |
| 2026-05-13 |
shaofeiqi | opencl: add q5_0 and q5_1 MoE for Adreno (#22985) |
commit | commitdiff | tree |
| 2026-05-13 |
Pascal | server, webui: accept continue_final_message flag for... |
commit | commitdiff | tree |
| 2026-05-13 |
lhez | opencl: fix crash when warming up MoE on Adreno (#22876) |
commit | commitdiff | tree |
| 2026-05-13 |
Masashi Yoshimura | flush the gpu profile timestamp before the queryset... |
commit | commitdiff | tree |
| 2026-05-13 |
Aleksander... | webui: Deduplicate model aliases in data + handle singl... |
commit | commitdiff | tree |
| 2026-05-13 |
Pascal | webui: preserve system message on edit cancel (#22911) |
commit | commitdiff | tree |
| 2026-05-13 |
Ravi Panchumarthy | docs : Update OPENVINO.md (#22959) |
commit | commitdiff | tree |
| 2026-05-13 |
Max Krasnyansky | hexagon: add unary tanh op (#22999) |
commit | commitdiff | tree |
| 2026-05-13 |
Xuan-Son Nguyen | download: do not exit() on error (#23008) |
commit | commitdiff | tree |
| 2026-05-13 |
Pascal | server, webui: support continue generation on reasoning... |
commit | commitdiff | tree |
| 2026-05-13 |
Xuan-Son Nguyen | ci: validate model naming convention (#22680) |
commit | commitdiff | tree |
| 2026-05-13 |
Georgi Gerganov | spec : update CLI arguments for better consistency... |
commit | commitdiff | tree |
| 2026-05-13 |
Sigbjørn Skjæret | llama-eval : enable type check (#22988) |
commit | commitdiff | tree |
| 2026-05-13 |
Sachin Sharma | ggml-zendnn : adaptive fallback to CPU backend for... |
commit | commitdiff | tree |
| 2026-05-13 |
Trivikram Reddy | hexagon: eliminate scalar VTCM loads via HVX splat... |
commit | commitdiff | tree |
| 2026-05-12 |
yzyyzyhhh | opencl: add opt-in Adreno xmem F16xF32 GEMM for prefill... |
commit | commitdiff | tree |
| 2026-05-12 |
fredzillman | convert : fix Pixtral 12B --mistral-format conversion... |
commit | commitdiff | tree |
| 2026-05-12 |
Aleksander... | webui: Fix Chat Screen Form box disappearing + autoscro... |
commit | commitdiff | tree |
| 2026-05-12 |
Xuan-Son Nguyen | mtmd, server, common: expose modalities to /v1/models... |
commit | commitdiff | tree |
| 2026-05-12 |
Masashi Yoshimura | ggml-webgpu: Enables running gpt-oss-20b (#22906) |
commit | commitdiff | tree |
| 2026-05-12 |
Chen Yuan | ggml-webgpu: address precision issues for multimodal... |
commit | commitdiff | tree |
| next |
overview / pkg / ggml / sources / llama.cpp / shortlog