git.djapps.eu Git - pkg/ggml/sources/llama.cpp/history

]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/history

overview / pkg / ggml / sources / llama.cpp / history

2026-08-10	Pascal	ui: UI/chat form follow ups (#26743)	\| commitdiff
2026-08-10	Sigbjørn Skjæret	ci : don't specify python version in server-sanitize...	\| commitdiff
2026-08-10	Pascal	server: add more tool isolation support (ssh remote...	\| commitdiff
2026-08-10	Pedro Cuenca	model: Muse Glimmer Support (#26841)	\| commitdiff
2026-08-10	Guido Imperiale	chat : Align Laguna-S-2.1 chat template to huggingface...	\| commitdiff
2026-08-10	Pascal	vendor: sync subprocess.h and drop local patches (...	\| commitdiff
2026-08-10	Pedro Cuenca	llama: Restore quantization of mmprojs (#26818)	\| commitdiff
2026-08-10	shivamkumard-ctrl	ci: Add support for CUDA 13.4 ARM64 builds for Windows...	\| commitdiff
2026-08-10	Ruixiang Wang	model: add MTP support for Nemotron model (#26725)	\| commitdiff
2026-08-10	Alessandro de Oliv...	vendor : update cpp-httplib to 0.53.0 (#26821)	\| commitdiff
2026-08-10	Bar Haim	model : Granite-Switch Architecture (#25107)	\| commitdiff
2026-08-10	Georgi Gerganov	readme : remove dev branches (#26832)	\| commitdiff
2026-08-10	Aleksander Grygier	ui: Linting & Formatting scripts (#26819)	\| commitdiff
2026-08-10	Pascal	server: gate the docker tools runtime tests on a real...	\| commitdiff
2026-08-10	Caleb DeLeeuw	model-saver : fix expert shared/chunk FFN length key...	\| commitdiff
2026-08-10	Eve	ci: fix the ctest sanitize runs (#26593)	\| commitdiff
2026-08-10	Masashi Yoshimura	ggml-webgpu : refactor several wgsl files and simplify...	\| commitdiff
2026-08-09	Pascal	ui: degrade the working directory picker when file...	\| commitdiff
2026-08-09	Xuan-Son Nguyen	ci: add pr-draft-label (#26801)	\| commitdiff
2026-08-09	Hao-Chen2337	ggml-cpu : fix missing Q5_0 dispatch in SpaceMiT backen...	\| commitdiff
2026-08-09	Aaron Teo	ci: rm `GGML_HIP_ROCWMMA_FATTN` (#26760)	\| commitdiff
2026-08-08	Pascal	server: report the isolate working directory from get_i...	\| commitdiff
2026-08-08	Rafail Giavrimis	CUDA: fuse rms_norm + mul + rope (+ view + set_rows...	\| commitdiff
2026-08-08	Pascal	server, ui: only offer a working directory when a tool...	\| commitdiff
2026-08-08	Xuan-Son Nguyen	server: add initial tool isolation support (via docker...	\| commitdiff
2026-08-08	Rafail Giavrimis	CUDA: fix thread/block count in quantized cpy kernel...	\| commitdiff
2026-08-07	Pascal	tts: account for the vocoder pass in the timings line...	\| commitdiff
2026-08-07	Aleksander Grygier	allozaur/feat/chat form contenteditable (#26717)	\| commitdiff
2026-08-07	Georgi Gerganov	tests : speed-up server test suite 3x (#26734)	\| commitdiff
2026-08-07	Aleksander Grygier	allozaur/feat/chat slash commands (#26716)	\| commitdiff
2026-08-07	Titaniumtown	sycl: coalesce the ssm_conv window loads (#26612)	\| commitdiff
2026-08-07	robertomeroni	metal : fix NORM/RMS_NORM for row lengths that leave...	\| commitdiff
2026-08-07	Aleksander Grygier	ui: Filesystem `@mentions` for Chat Form (#26715)	\| commitdiff
2026-08-07	Xuan-Son Nguyen	mtmd: fix longest_edge ignoring min/max pixels (#26638)	\| commitdiff
2026-08-07	Georgi Gerganov	sync : ggml	\| commitdiff
2026-08-07	Georgi Gerganov	ggml : bump version to 0.19.0 (ggml/1581)	\| commitdiff
2026-08-07	Daniel Bevenius	server : clarify comment in eval_llama_cmpl_schema...	\| commitdiff
2026-08-07	Emanuil Rusev	webui: load the model selected via ?model= when ?load...	\| commitdiff
2026-08-07	Niklas Wenzel	ui: set npm `min-release-age` to protect against supply...	\| commitdiff
2026-08-07	Xuan-Son Nguyen	server: (router) add LRU scheduler (#26572)	\| commitdiff
2026-08-07	Xuan-Son Nguyen	server: (router) do not evict busy models (#26567)	\| commitdiff
2026-08-07	Pascal	mtmd: stop feeding the text stream again during Qwen3...	\| commitdiff
2026-08-07	Kilian Hu	ggml : add aarch64 HWCAP fallbacks and fix fp16 variant...	\| commitdiff
2026-08-07	Pascal	ui: read model modalities from the router model list...	\| commitdiff
2026-08-07	Masato Nakasaka	Mitigate crashing issue on Windows MSYS2 UCRT64 environ...	\| commitdiff
2026-08-07	Chris Lee	sycl: fix UE4M3 parsing (#25608)	\| commitdiff
2026-08-07	Titaniumtown	sycl: *glu flat path (#26354)	\| commitdiff
2026-08-07	Neo Zhang	sycl : Support DSv4 OPs: LIGHTNING_INDEXER,DSV4_HC_COMB...	\| commitdiff
2026-08-07	Neo Zhang	sycl : update guide Q&A and script for device setting...	\| commitdiff
2026-08-07	Neo Zhang	sycl : fix error Error OP FLASH_ATTN_EXT on arc770...	\| commitdiff
2026-08-07	Neo Zhang	sycl : enhance OP set_rows to support all missed data...	\| commitdiff
2026-08-07	David Friehs	cuda: fix warnings for unused variable/function (#26688)	\| commitdiff
2026-08-07	Niklas Wenzel	ci: abort if build requirements are missing (#26368)	\| commitdiff
2026-08-07	JamePeng	metal : avoid `threadgroup` matrix array instantiation...	\| commitdiff
2026-08-06	Xuan-Son Nguyen	mtmd: add chunk save/load function (#26645)	\| commitdiff
2026-08-06	Xuan-Son Nguyen	server: fix empty response for /cors-proxy (#26656)	\| commitdiff
2026-08-06	Sigbjørn Skjæret	convert : fix DeepseekV4 rope parameters with transform...	\| commitdiff
2026-08-06	Georgi Gerganov	model-loader : fix quantized reshaped tensor strides...	\| commitdiff
2026-08-06	Csaba Kecskemeti	convert : accept "ExaoneMoeForCausalLM" arch spelling...	\| commitdiff
2026-08-06	Jim Wu	ci : onboard AMD ROCm CI with gfx1151 fixes (#26544)	\| commitdiff
2026-08-06	Daniel Bevenius	model-conversion : add --model-name to conversion scrip...	\| commitdiff
2026-08-06	Ruben Ortlam	vulkan: fix submission batching size, add debug tools...	\| commitdiff
2026-08-05	Pascal	mtmd/ggml: add ggml_build_forward_order (#26649)	\| commitdiff
2026-08-05	Pascal	server: harden the file_glob_search directory walk...	\| commitdiff
2026-08-05	Niklas Wenzel	tests: re-enable MiniMax M3 in `test-llama-archs` ...	\| commitdiff
2026-08-05	Saba Fallah	mtmd: Unlimited-OCR fix max_tiles, setting in converter...	\| commitdiff
2026-08-05	Aldehir Rojas	grammar : degrade max repetition >= 2000 to unbounded...	\| commitdiff
2026-08-05	Xuan-Son Nguyen	mtmd: support multi-row batching for deepseek-ocr ...	\| commitdiff
2026-08-05	Sergey Malinin	fit: Fix memory allocation for MTP layers (#26605)	\| commitdiff
2026-08-05	Xuan-Son Nguyen	security : clarify about AI-generated reports (#26579)	\| commitdiff
2026-08-05	Bhavik Sharda	server: Adding spec-decode counters to /metrics endpoin...	\| commitdiff
2026-08-05	Andreas Krebbel	convert: Add endianness conversion for Q1 and TQ2 quant...	\| commitdiff
2026-08-05	Xuan-Son Nguyen	vendor : apply patches for subprocess.h (#26606)	\| commitdiff
2026-08-05	Aleksander Grygier	ui: show generation statistics by default in chat setti...	\| commitdiff
2026-08-05	Niklas Wenzel	build : remove GGML_METAL_USE_BF16 from all build scrip...	\| commitdiff
2026-08-05	Aleksander Grygier	ui: Update vulnerable packages + cleanup Storybook...	\| commitdiff
2026-08-04	Evan Huus	Prefer npm ci over install for security (#26601)	\| commitdiff
2026-08-04	Pascal	server: decode Windows OEM output to UTF-8 in built...	\| commitdiff
2026-08-04	Abhinay Krishna	mtmd: correcting duplicate empty audio chunks for short...	\| commitdiff
2026-08-04	Oliver Simons	sampler : remove "full-context windows" from history...	\| commitdiff
2026-08-04	Guilherme Quintino	gguf-split: Add option to delete split parts during...	\| commitdiff
2026-08-04	Aleksander Grygier	ui: CWD for agent (#26518)	\| commitdiff
2026-08-04	Xuan-Son Nguyen	mtmd: support Qwen3-TTS (note: breaking change to llama...	\| commitdiff
2026-08-04	Georgi Gerganov	models : fix dflash wo_a reshape on load (#26577)	\| commitdiff
2026-08-04	Niklas Wenzel	ci: fix pre-built binaries no longer working on macOS...	\| commitdiff
2026-08-04	Daniel Bevenius	speculative : refactor enabled configs common_speculati...	\| commitdiff
2026-08-04	hcl	gguf-py: validate n_dims and guard against uint64 overf...	\| commitdiff
2026-08-04	Georgi Gerganov	sync : ggml	\| commitdiff
2026-08-04	Georgi Gerganov	ggml : bump version to 0.18.1 (ggml/1578)	\| commitdiff
2026-08-04	Angel Galindo	convert : add missing return after setting tekken vocab...	\| commitdiff
2026-08-04	Pranav Uttarkar	vulkan backend ops: implemented GATED_LINEAR_ATTN ...	\| commitdiff
2026-08-04	Sigbjørn Skjæret	vocab : validate plamo2 byte tokens (#26511)	\| commitdiff
2026-08-04	Caleb DeLeeuw	convert : import bytes_to_unicode from convert_slow_tok...	\| commitdiff
2026-08-04	Georgi Gerganov	model : allow reshape of tensors during load (#26531)	\| commitdiff
2026-08-04	Oliver Simons	llama : move n_vocab from llama_sampler_data to penalty...	\| commitdiff
2026-08-04	Eve	ci: fix vulkan llvmpipe runs (#26533)	\| commitdiff
2026-08-04	Titaniumtown	sycl: parallelize the non-contiguous concat kernel...	\| commitdiff
2026-08-04	Ozymandias_EBON	Extended SYCL oneDNN SDPA to non-FP16 KV caches (Q4_0...	\| commitdiff
2026-08-03	Thiago Padilha	chat : add new template for DeepSeek V4 Flash 0731...	\| commitdiff
2026-08-03	Alessandro de Oliv...	vendor : update cpp-httplib to 0.52.0 (#26485)	\| commitdiff
next

Packaging of ggml-org/llama.cpp

RSS Atom