ik_llama.cpp.git - Unnamed repository; edit this file 'description' to name the repository.

Age	Commit message (Expand)	Author
2025-02-10	Load all MoE experts during warmup and make warmup 1 token (#198)	saood06
2025-02-09	Add optional MLA (#188)	Kawrakow
2025-01-30	Faster Q4_K_R4 and Q5_K_R4 on AVX2/Zen4 (#182)	Kawrakow
2025-01-29	Various (#181)	Kawrakow
2024-12-17	Slightly better matrix x vector on Zen4/AVX2 for iq2_k_r4, iq3_k_r4, iq4_k_r4...	Kawrakow
2024-12-17	Be able to repack tensors at run time (#147)	Kawrakow
2024-10-02	Adding Q6_0 (#77)	Kawrakow
2024-09-05	Zen4 Flash Attention - bf16 support (#38)	Kawrakow
2024-08-20	Fused soft cap and SIMD-ified GeLU (#9)	Kawrakow
2024-08-12	Merge mainline - Aug 12 2024 (#17)	Kawrakow
2024-07-27	Merge mainline llama.cpp (#3)	Kawrakow
2024-06-14	llama-bench : fix RPC indication (#7936)	Radoslav Gerganov
2024-06-13	move BLAS to a separate backend (#6210)	slaren
2024-06-11	llama-bench: more compact markdown tables (#7879)	Johannes Gäßler
2024-06-04	common : refactor cli arg parsing (#7675)	Georgi Gerganov
2024-06-04	ggml : remove OpenCL (#7735)	Georgi Gerganov
2024-06-04	llama-bench : allow using a different printer for stderr with -oe (#7722)	slaren
2024-05-29	llama-bench : add support for the RPC backend (#7435)	Radoslav Gerganov
2024-05-22	common : normalize naming style (#7462)	Georgi Gerganov
2024-05-22	phi3 : duplicate rope factors in each layer (#7447)	slaren
2024-05-10	llama-bench : add pp+tg test type (#7199)	slaren
2024-05-05	Adding support for the --numa argument for llama-bench. (#7080)	kunnis
2024-04-30	ggml : add Flash Attention (#5021)	Georgi Gerganov
2024-04-16	ggml : add llamafile sgemm (#6414)	Justine Tunney
2024-03-26	cuda : rename build flag to LLAMA_CUDA (#6299)	slaren
2024-03-21	Add ability to use Q5_0, Q5_1, and IQ4_NL for quantized K cache (#6183)	Kawrakow
2024-03-18	backend : offload large batches to GPU (#6083)	slaren
2024-03-15	llama-bench : use random tokens to improve accuracy with mixtral (#6069)	slaren
2024-03-14	gguf : fix resource leaks (#6061)	Steve Grubb
2024-03-13	llama : add pipeline parallelism support (#6017)	slaren
2024-03-07	llama-bench : add embeddings option (#5924)	Georgi Gerganov
2024-03-02	Support multiple GPUs (split mode) on SYCL backend (#5806)	Neo Zhang Jianyu
2024-03-01	llama : cleanup unused mmq flags (#5772)	Pierrick Hymbert
2024-02-25	code : normalize enum names (#5697)	Georgi Gerganov
2024-02-16	ggml : add numa options (#5377)	bmwl
2024-02-03	refactor : switch to emplace_back to avoid extra object (#5291)	Michael Klimenko
2024-02-01	add --no-mmap in llama-bench (#5257)	Neo Zhang Jianyu
2024-01-31	llama : remove LLAMA_MAX_DEVICES and LLAMA_SUPPORTS_GPU_OFFLOAD (#5240)	Georgi Gerganov
2024-01-30	kompute : llama-bench support and ggml_cpu_has_kompute() (#5226)	Jared Van Bortel
2024-01-28	ggml : add Vulkan backend (#2059)	0cc4m
2024-01-12	llama : ggml-backend integration (#4766)	slaren
2024-01-07	llama-bench : add no-kv-offload parameter (#4812)	slaren
2023-12-07	llama : per-layer KV cache + quantum K cache (#4309)	Georgi Gerganov
2023-11-02	build : link against build info instead of compiling against it (#3879)	cebtenzzre
2023-10-29	Extend llama_kv_cache_seq_rm to allow matching any sequence (#3843)	Kerfuffle
2023-10-23	llama : remove token functions with `context` args in favor of `model` (#3720)	Marcus Dunn
2023-09-28	build : enable more non-default compiler warnings (#3200)	Cebtenzzre
2023-09-28	llama.cpp : split llama_context_params into model and context params (#3301)	slaren
2023-09-28	llama : custom attention mask + parallel decoding + no context swaps (#3228)	Georgi Gerganov
2023-09-27	metal : reusing llama.cpp logging (#3152)	Rickard Hallerbäck