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

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

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

2025-09-01	Jie Fu (傅杰)	docs : add Hunyuan to models section (#15707)	commit \| commitdiff \| tree
2025-09-01	Akarshan Biswas	CUDA: fix build error from ambiguous __half conversions...	commit \| commitdiff \| tree
2025-09-01	hipudding	CANN: Optimize MUL_MAT_ID (#15658)	commit \| commitdiff \| tree
2025-09-01	hipudding	CANN: fix RoPE cache issue on multi-device (#15629)	commit \| commitdiff \| tree
2025-08-31	Georgi Gerganov	sampling : optimize samplers by reusing bucket sort...	commit \| commitdiff \| tree
2025-08-31	Georgi Gerganov	server : enable /slots by default and make it secure...	commit \| commitdiff \| tree
2025-08-31	Georgi Gerganov	metal : fix checks for available FA kernels (#15700)	commit \| commitdiff \| tree
2025-08-31	Diego Devesa	llama : fix fattn reserve call n_seqs parameter (#15699)	commit \| commitdiff \| tree
2025-08-31	Diego Devesa	llama : separate compute buffer reserve from fattn...	commit \| commitdiff \| tree
2025-08-31	Sigbjørn Skjæret	ci : explicitly set fa off or on (#15692)	commit \| commitdiff \| tree
2025-08-31	Jeff Bolz	vulkan: handle large sizes for get_rows (#15686)	commit \| commitdiff \| tree
2025-08-31	Jeff Bolz	vulkan: mul_mat_id coopmat2 optimizations (#15546)	commit \| commitdiff \| tree
2025-08-31	Daniel Bevenius	vulkan : remove unused portability_enumeration_ext...	commit \| commitdiff \| tree
2025-08-31	Jeff Bolz	vulkan: Allow fallback to sysmem memory when vidmem...	commit \| commitdiff \| tree
2025-08-31	Jeff Bolz	vulkan: clamp matmul and FA results to the max finite...	commit \| commitdiff \| tree
2025-08-30	Charles Xu	ggml: update kleidiai to v1.13.0 (#15663)	commit \| commitdiff \| tree
2025-08-30	Diego Devesa	Update build.md to remove MSVC arm64 notes (#15684)	commit \| commitdiff \| tree
2025-08-30	Johannes Gäßler	llama: use FA + max. GPU layers by default (#15434)	commit \| commitdiff \| tree
2025-08-30	Johannes Gäßler	CUDA: use FP32 arithmetic for conv2d (#15683)	commit \| commitdiff \| tree
2025-08-30	Jeff Bolz	vulkan: Skip syncing for prealloc_y when it is reused...	commit \| commitdiff \| tree
2025-08-30	Chenguang Li	CANN: FIx compiler warnings (#15661)	commit \| commitdiff \| tree
2025-08-29	Sergey Alirzaev	server : removed obsolete doc (#15670)	commit \| commitdiff \| tree
2025-08-29	Johannes Gäßler	scripts: strip "AMD Instinct" from GPU name (#15668)	commit \| commitdiff \| tree
2025-08-29	ExtReMLapin	server : add documentation for `parallel_tool_calls...	commit \| commitdiff \| tree
2025-08-29	Aman Gupta	CUDA: fix bug in rms_norm fusion (#15660)	commit \| commitdiff \| tree
2025-08-29	Piotr Wilkin...	chat : Seed OSS thinking + tool call support (#15552)	commit \| commitdiff \| tree
2025-08-29	Aman Gupta	CUDA: fuse adds, fuse add with rms norm (#15631)	commit \| commitdiff \| tree
2025-08-29	Gabe Goodhart	nvidia nemotron nano v2 (nemotronh) (#15507)	commit \| commitdiff \| tree
2025-08-28	Gabe Goodhart	fix: Compute the full sum in llama-eval-callback, not...	commit \| commitdiff \| tree
2025-08-28	mnehete32	CUDA: add conv2d (#15635)	commit \| commitdiff \| tree
2025-08-28	Aaron Teo	ggml-cpu: fix invalid hsum build in debug s390x (#15634)	commit \| commitdiff \| tree
2025-08-28	compilade	ggml : fix SSM_SCAN for n_groups > 1 (#15625)	commit \| commitdiff \| tree
2025-08-28	Georgi Gerganov	kv-cache : fix find_slot to not search for continuous...	commit \| commitdiff \| tree
2025-08-28	Sigbjørn Skjæret	model : jina-embeddings-v3 support (#13693)	commit \| commitdiff \| tree
2025-08-28	Aman Gupta	scripts: add sqlite3 check for compare-commits.sh ...	commit \| commitdiff \| tree
2025-08-28	Georgi Gerganov	kv-cache : remove LLAMA_SET_ROWS checks (#15505)	commit \| commitdiff \| tree
2025-08-28	Aleksei Nikiforov	gguf-py: byteswapping improvements (#12851)	commit \| commitdiff \| tree
2025-08-28	Joshua Cogliati	cli : change log to warning to explain reason for stopp...	commit \| commitdiff \| tree
2025-08-28	Daniel Bevenius	model-conversion : add mmproj conversion target (#15628)	commit \| commitdiff \| tree
2025-08-28	matiaslin	cuda: Add cublasLt_static linking when GGML_STATIC...	commit \| commitdiff \| tree
2025-08-27	Johannes Gäßler	server: higher timeout for tests (#15621)	commit \| commitdiff \| tree
2025-08-27	Georgi Gerganov	presets : add qwen3-30B-a3b FIM (#15616)	commit \| commitdiff \| tree
2025-08-27	uvos	HIP: Enable support for ggml_backend_cuda_register_host...	commit \| commitdiff \| tree
2025-08-27	Georgi Gerganov	kv-cache : better estimate of n_kv for multi-sequence...	commit \| commitdiff \| tree
2025-08-27	Chenguang Li	CANN: refactor mask handling and improve performance...	commit \| commitdiff \| tree
2025-08-27	xctan	ggml-cpu : add basic RVV support for vector f32 ops...	commit \| commitdiff \| tree
2025-08-27	Daniel Bevenius	common : add -m to bash completion for --model [no...	commit \| commitdiff \| tree
2025-08-27	rmatif	OpenCL: add fused group_norm/norm, mul, add (#15314)	commit \| commitdiff \| tree
2025-08-26	Diego Devesa	tests : fix test-opt with GGML_BACKEND_DL (#15599)	commit \| commitdiff \| tree
2025-08-26	Akarshan Biswas	SYCL: fix rms_norm_mul_add for tensor dim not a multipl...	commit \| commitdiff \| tree
2025-08-26	fidoriel	mtmd : fix mtmd ios build (#15579)	commit \| commitdiff \| tree
2025-08-26	Eve	tests: add performance test for mul mat id (#15543)	commit \| commitdiff \| tree
2025-08-26	shalinib-ibm	llamafile: PowerPC Sgemm Optimization (#15558)	commit \| commitdiff \| tree
2025-08-26	Georgi Gerganov	graph : fix assert in memory-less build_attn (#15590)	commit \| commitdiff \| tree
2025-08-26	Daniel Bevenius	model-conversion : add qat-q4 quantization targets...	commit \| commitdiff \| tree
2025-08-26	Johannes Gäßler	CUDA: return -1 for nonexistent compiled arch (#15587)	commit \| commitdiff \| tree
2025-08-26	Georgi Gerganov	metal : optimize FA vec for large sequences and BS...	commit \| commitdiff \| tree
2025-08-26	Xuan-Son Nguyen	mtmd : support Kimi VL model (#15458)	commit \| commitdiff \| tree
2025-08-26	Georgi Gerganov	context : print graph stats for memory-less contexts...	commit \| commitdiff \| tree
2025-08-26	Georgi Gerganov	metal : improve `MUL_MAT_ID` (#15541)	commit \| commitdiff \| tree
2025-08-26	tc-mb	model : support MiniCPM-V 4.5 (#15575)	commit \| commitdiff \| tree
2025-08-26	Sigbjørn Skjæret	gguf-py : remove erroneous FFN_GATE entry (#15583)	commit \| commitdiff \| tree
2025-08-26	Sigbjørn Skjæret	metal : remove contiguous assertion for src0 in IM2COL...	commit \| commitdiff \| tree
2025-08-26	Yoshi_likes_e4	Add a warning for special devices (#15563)	commit \| commitdiff \| tree
2025-08-26	Jeff Bolz	vulkan: Remove splitting for mul_mat_id (#15568)	commit \| commitdiff \| tree
2025-08-25	Qeeweew	CUDA: Accelerate MXFP4 table lookup using `__byte_perm...	commit \| commitdiff \| tree
2025-08-25	lhez	opencl: fix support ops condition for `rms_norm` (...	commit \| commitdiff \| tree
2025-08-25	Ruben Ortlam	vulkan: fix min subgroup 16 condition for mmid subgroup...	commit \| commitdiff \| tree
2025-08-25	Jeff Bolz	tests: Generate unique input values for count_equal...	commit \| commitdiff \| tree
2025-08-25	Ihar Hrachyshka	metal: fix regression when no metal devices are present...	commit \| commitdiff \| tree
2025-08-25	Johannes Gäßler	CUDA: MoE helper in device code, better tile sizes...	commit \| commitdiff \| tree
2025-08-25	Daniel Bevenius	model-conversion : set pooling type to none in logits...	commit \| commitdiff \| tree
2025-08-25	Daniel Bevenius	model-conversion : add model card template for embeddin...	commit \| commitdiff \| tree
2025-08-25	Georgi Gerganov	batched-bench : fix unified KV cache handling + pp...	commit \| commitdiff \| tree
2025-08-25	Weizhao Ouyang	convert : update Ernie 4.5 dense architecture name...	commit \| commitdiff \| tree
2025-08-25	Georgi Gerganov	metal : add FA kernels for HS=40 (#15559)	commit \| commitdiff \| tree
2025-08-25	RunningLeon	convert : support interns1-mini (#15412)	commit \| commitdiff \| tree
2025-08-25	Chenguang Li	CANN: ROPE cache sin/cos repeat (#15501)	commit \| commitdiff \| tree
2025-08-24	Ruben Ortlam	vulkan: apply MUL_MAT_ID subgroup optimization to non...	commit \| commitdiff \| tree
2025-08-24	Georgi Gerganov	kv-cache : support layer reuse (#15504)	commit \| commitdiff \| tree
2025-08-24	Jeff Bolz	vulkan: Support FA with any multiple of 8 head sizes...	commit \| commitdiff \| tree
2025-08-24	Ruben Ortlam	vulkan: enable Conv2D for Apple after MoltenVK fixed...	commit \| commitdiff \| tree
2025-08-24	Jeff Bolz	vulkan: workaround MoltenVK compile failure in multi_ad...	commit \| commitdiff \| tree
2025-08-23	Johannes Gäßler	CUDA: fix half2 -> half conversion for HIP (#15529)	commit \| commitdiff \| tree
2025-08-23	Jeff Bolz	vulkan: optimize rms_norm, and allow the work to spread...	commit \| commitdiff \| tree
2025-08-23	Piotr Wilkin...	model : add support for Seed-OSS (#15490)	commit \| commitdiff \| tree
2025-08-23	Johannes Gäßler	scripts: fix compare-llama-bench.py (#15521)	commit \| commitdiff \| tree
2025-08-23	LaffeyNyaa	chat : fix debug build assertion in trim function ...	commit \| commitdiff \| tree
2025-08-23	Jeff Bolz	vulkan: Rewrite synchronization to allow some overlap...	commit \| commitdiff \| tree
2025-08-23	R0CKSTAR	vulkan.Dockerfile: install vulkan SDK using tarball...	commit \| commitdiff \| tree
2025-08-23	Acly	vulkan : support ggml_mean (#15393)	commit \| commitdiff \| tree
2025-08-23	Jeff Bolz	vulkan: optimize mul_mat_id loading row ids into shared...	commit \| commitdiff \| tree
2025-08-22	Johannes Gäßler	test-opt: allow slight inprecision (#15503)	commit \| commitdiff \| tree
2025-08-22	Reese Levine	ggml WebGPU: add support for quantization types (#15440)	commit \| commitdiff \| tree
2025-08-22	Aldehir Rojas	model : gpt-oss add response_format support (#15494)	commit \| commitdiff \| tree
2025-08-22	rmatif	ggml: add `conv3d` op (#15182)	commit \| commitdiff \| tree
2025-08-22	Yavor Ivanov	cuda : add Pad Reflect 1D support (#14659)	commit \| commitdiff \| tree
2025-08-22	Georgi Gerganov	llama : remove KV cache defragmentation logic (#15473)	commit \| commitdiff \| tree
2025-08-22	Aaron Teo	ggml-cpu: Support Q5_0 and Q5_1 on s390x (#15486)	commit \| commitdiff \| tree
2025-08-22	65a	server : Support multimodal completion and embeddings...	commit \| commitdiff \| tree
next

Packaging of ggml-org/llama.cpp

RSS Atom