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

2025-05-15	Xuan-Son Nguyen	webui : handle PDF input (as text or image) + convert...	commit \| commitdiff \| tree
2025-05-15	Piotr Wilkin...	server : proper error handling for missing elements...	commit \| commitdiff \| tree
2025-05-15	Georgi Gerganov	bench : handle decode errors (#13548)	commit \| commitdiff \| tree
2025-05-15	Olivier Chafik	`server`: inject date_string in llama 3.x template...	commit \| commitdiff \| tree
2025-05-14	Georgi Gerganov	kv-cache : fix out-of-bounds view during reserve graph...	commit \| commitdiff \| tree
2025-05-14	Yibo Cai	arm64: optimize q6_k_q8_k kernel with i8mm (#13519)	commit \| commitdiff \| tree
2025-05-14	Olivier Chafik	`common`: add partial regex support (#12808)	commit \| commitdiff \| tree
2025-05-14	Sigbjørn Skjæret	editorconfig : fix trailing whitespace from #13542...	commit \| commitdiff \| tree
2025-05-14	Gilad S.	fix: crash when calling `llama_state_get_size` on a...	commit \| commitdiff \| tree
2025-05-14	Johannes Gäßler	CUDA: fix crash on large batch size for quant. MoE...	commit \| commitdiff \| tree
2025-05-14	Diego Devesa	llama : fix quantize with dl backends (#13539)	commit \| commitdiff \| tree
2025-05-14	Johannes Gäßler	CUDA: faster Deepseek FA, add Turing support (#13435)	commit \| commitdiff \| tree
2025-05-14	Gabe Goodhart	fix: Move build_inp_pos to the top of the graph section...	commit \| commitdiff \| tree
2025-05-14	Georgi Gerganov	server : passthrough the /models endpoint during loadin...	commit \| commitdiff \| tree
2025-05-14	Xuan-Son Nguyen	server : fix cache_tokens bug with no cache_prompt...	commit \| commitdiff \| tree
2025-05-14	bandoti	cmake: simplify vulkan shader test logic (#13263)	commit \| commitdiff \| tree
2025-05-14	Jeff Bolz	vulkan: KHR_coopmat flash attention (#13506)	commit \| commitdiff \| tree
2025-05-14	Xuan-Son Nguyen	webui : use fflate for more deterministic gzip compress...	commit \| commitdiff \| tree
2025-05-14	Luca Stefani	webui: Allow pasting file from clipboard (#13526)	commit \| commitdiff \| tree
2025-05-14	ddpasa	docs: Update link to ggml-org in multimodal.md (#13513)	commit \| commitdiff \| tree
2025-05-14	Sigbjørn Skjæret	scripts : fix compare-llama-bench.py show parameter...	commit \| commitdiff \| tree
2025-05-14	Jeff Bolz	vulkan: workaround FA compile failures on macos (#13517)	commit \| commitdiff \| tree
2025-05-13	Ed Addario	quantize : improve tensor-type pattern matching (#13033)	commit \| commitdiff \| tree
2025-05-13	Xuan-Son Nguyen	clip : clip.h become private API (⚠️ breaking change...	commit \| commitdiff \| tree
2025-05-13	Georgi Gerganov	metal : use FA-vec kernel up to batch size 20 (#13496)	commit \| commitdiff \| tree
2025-05-13	Georgi Gerganov	metal : optimize multi-sequence FA vec kernel (#13493)	commit \| commitdiff \| tree
2025-05-13	Dan Johansson	ggml-cpu: Update KleidiAI to v1.6 and fix include direc...	commit \| commitdiff \| tree
2025-05-13	Georgi Gerganov	batched-bench : fix pp batch contents (#13492)	commit \| commitdiff \| tree
2025-05-13	Xuan-Son Nguyen	mtmd : remove libllava, remove clip-quantize-cli (...	commit \| commitdiff \| tree
2025-05-13	Sigbjørn Skjæret	scripts : support arbitrary input file formats in compa...	commit \| commitdiff \| tree
2025-05-13	Gabe Goodhart	model : Granite MoE shared (#13269)	commit \| commitdiff \| tree
2025-05-13	Georgi Gerganov	sync : ggml	commit \| commitdiff \| tree
2025-05-12	Diego Devesa	llama-bench : add defrag-thold, check for invalid range...	commit \| commitdiff \| tree
2025-05-12	lhez	opencl: remove unnecessary assert for `add` (#13257)	commit \| commitdiff \| tree
2025-05-12	Xuan-Son Nguyen	clip : cap max image size 1024 for qwen vl model (...	commit \| commitdiff \| tree
2025-05-12	Johannes Gäßler	llama/ggml: add LLM training support (#10544)	commit \| commitdiff \| tree
2025-05-12	Georgi Gerganov	context : fix state io for memory-less contexts (#13470)	commit \| commitdiff \| tree
2025-05-12	Anudit Nagar	server : allow content to be null in oaicompat_completi...	commit \| commitdiff \| tree
2025-05-12	Diego Devesa	llama-bench : accept ranges for integer parameters...	commit \| commitdiff \| tree
2025-05-12	Dan Johansson	ggml-cpu: Integrate fp32=bf16xbf16 SME KleidiAI kernel...	commit \| commitdiff \| tree
2025-05-12	Johannes Gäßler	CUDA: fix misaligned synchronization in FA (#13469)	commit \| commitdiff \| tree
2025-05-12	Xuan-Son Nguyen	ggml : add mrope kernel for metal (#13457)	commit \| commitdiff \| tree
2025-05-12	Atharva Dubey	enable dpcpp nightly builds with libraries (#13406)	commit \| commitdiff \| tree
2025-05-11	City	mtmd : Use RMS norm for InternVL 3 38B and 78B mmproj...	commit \| commitdiff \| tree
2025-05-11	Anthony Umfer	tools : fix uninitialized llama_batch in server (#13436)	commit \| commitdiff \| tree
2025-05-11	Sigbjørn Skjæret	scripts : exit compare-llama-bench.py gracefully when...	commit \| commitdiff \| tree
2025-05-11	Johannes Gäßler	CUDA: fix crash with partial offloading of MoE (#13439)	commit \| commitdiff \| tree
2025-05-11	David Huang	Add `--no-op-offload` to improve `-ot` pp perf in MoE...	commit \| commitdiff \| tree
2025-05-11	City	mtmd : support InternVL 3 38B and 78B mmproj (#13443)	commit \| commitdiff \| tree
2025-05-11	Xuan-Son Nguyen	mtmd : move helpers to dedicated file (#13442)	commit \| commitdiff \| tree
2025-05-10	Thomas Germer	docs : Fix typo in InternVL3 model name (#13440)	commit \| commitdiff \| tree
2025-05-10	Johannes Gäßler	CUDA: fix race conditions FlashAttention kernels (...	commit \| commitdiff \| tree
2025-05-10	Sigbjørn Skjæret	vocab : add ByteDance-Seed/Seed-Coder (#13423)	commit \| commitdiff \| tree
2025-05-10	Xuan-Son Nguyen	mtmd : add hard limit on image resolution for qwen2vl...	commit \| commitdiff \| tree
2025-05-10	Xuan-Son Nguyen	server : update docs (#13432)	commit \| commitdiff \| tree
2025-05-10	Sigbjørn Skjæret	llguidance : set tokenizer slices to default (#13424)	commit \| commitdiff \| tree
2025-05-10	Thammachart...	ci: free_disk_space flag enabled for intel variant...	commit \| commitdiff \| tree
2025-05-10	Xuan-Son Nguyen	mtmd : support InternVL 2.5 and 3 (#13422)	commit \| commitdiff \| tree
2025-05-10	Johannes Gäßler	CUDA: fix FlashAttention on Turing (#13415)	commit \| commitdiff \| tree
2025-05-10	Xuan-Son Nguyen	arg : add env var to control mmproj (#13416)	commit \| commitdiff \| tree
2025-05-10	Jeff Bolz	vulkan: scalar flash attention implementation (#13324)	commit \| commitdiff \| tree
2025-05-09	Helton Reis	chore(llguidance): use tagged version that does not...	commit \| commitdiff \| tree
2025-05-09	Xuan-Son Nguyen	server : vision support via libmtmd (#12898)	commit \| commitdiff \| tree
2025-05-09	Alberto Cabrera...	sycl : implementation of reordered Q4_0 MMVQ for Intel...	commit \| commitdiff \| tree
2025-05-09	Georgi Gerganov	metal : optimize MoE for large batches (#13388)	commit \| commitdiff \| tree
2025-05-09	Johannes Gäßler	CUDA: FA support for Deepseek (Ampere or newer) (#13306)	commit \| commitdiff \| tree
2025-05-09	Diego Devesa	llama : do not crash if there is no CPU backend (#13395)	commit \| commitdiff \| tree
2025-05-09	Johannes Gäßler	CUDA: fix crash on large batch size for MoE models...	commit \| commitdiff \| tree
2025-05-09	Bartowski	imatrix : Add --parse-special for enabling parsing...	commit \| commitdiff \| tree
2025-05-09	R0CKSTAR	llama-run: add support for downloading models from...	commit \| commitdiff \| tree
2025-05-09	Xuan-Son Nguyen	mtmd : fix batch_view for m-rope (#13397)	commit \| commitdiff \| tree
2025-05-09	Xuan-Son Nguyen	llama : one-off chat template fix for Mistral-Small...	commit \| commitdiff \| tree
2025-05-09	Radoslav Gerganov	rpc : add rpc_msg_set_tensor_hash_req (#13353)	commit \| commitdiff \| tree
2025-05-09	Jeff Bolz	vulkan: Allow up to 4096 elements for mul_mat_id row_id...	commit \| commitdiff \| tree
2025-05-09	Xuan-Son Nguyen	server : (webui) rename has_multimodal --> modalities...	commit \| commitdiff \| tree
2025-05-08	Diego Devesa	ci : limit write permission to only the release step... upstream/0.0.5318	commit \| commitdiff \| tree
2025-05-08	Matt Clayton	mtmd : Expose helper_decode_image_chunk (#13366)	commit \| commitdiff \| tree
2025-05-08	Xuan-Son Nguyen	server : (webui) fix a very small misalignment (#13387)	commit \| commitdiff \| tree
2025-05-08	Xuan-Son Nguyen	server : (webui) revamp the input area, plus many small...	commit \| commitdiff \| tree
2025-05-08	Sigbjørn Skjæret	convert : support rope_scaling type and rope_type ...	commit \| commitdiff \| tree
2025-05-08	welix	mtmd : fix the calculation of n_tokens for smolvlm...	commit \| commitdiff \| tree
2025-05-08	Georgi Gerganov	context : allow cache-less context for embeddings ...	commit \| commitdiff \| tree
2025-05-08	Georgi Gerganov	context : remove logits_all flag (#13284)	commit \| commitdiff \| tree
2025-05-08	Diego Devesa	ci : move release workflow to a separate file (#13362)	commit \| commitdiff \| tree
2025-05-08	Diego Devesa	llama : print size and type of overridden tensors ...	commit \| commitdiff \| tree
2025-05-08	Alberto Cabrera...	sycl: addressing non-contiguous src1 mul_mats (nc and...	commit \| commitdiff \| tree
2025-05-07	Diego Devesa	docker : disable arm64 and intel images (#13356)	commit \| commitdiff \| tree
2025-05-07	Georgi Gerganov	sync : ggml	commit \| commitdiff \| tree
2025-05-07	Daniel Bevenius	whisper: remove MSVC warnings pragmas (whisper/3090)	commit \| commitdiff \| tree
2025-05-07	Jared Tweed	cmake : removed stdc++fs (whisper/3097)	commit \| commitdiff \| tree
2025-05-07	Sigbjørn Skjæret	llama : deci : support ffn-free with attention (#13296)	commit \| commitdiff \| tree
2025-05-07	Ycros	common : Add a warning when we can't match samplers...	commit \| commitdiff \| tree
2025-05-07	R0CKSTAR	cuda : remove nrows_x in mul_mat_q_process_tile (#13325)	commit \| commitdiff \| tree
2025-05-07	Georgi Gerganov	examples : remove infill (#13283)	commit \| commitdiff \| tree
2025-05-07	piDack	llama : support tie embedding for chatglm models (...	commit \| commitdiff \| tree
2025-05-06	Johannes Gäßler	CUDA: mix virt/real CUDA archs for GGML_NATIVE=OFF...	commit \| commitdiff \| tree
2025-05-06	Xuan-Son Nguyen	clip : refactor graph builder (#13321)	commit \| commitdiff \| tree
2025-05-06	DocShotgun	sampling : make top_n_sigma no-op at <=0 or a single...	commit \| commitdiff \| tree
2025-05-06	oobabooga	sampling : don't consider -infinity values in top_n_sig...	commit \| commitdiff \| tree
2025-05-06	Diego Devesa	cmake : remove arm64 msvc presets (#13342)	commit \| commitdiff \| tree
next

Packaging of ggml-org/llama.cpp

RSS Atom