llama.cpp.git - llama.cpp

Age	Commit message (Expand)	Author
2023-07-28	llama : support more diverse tokenizers? (#2420)	eric8607242
2023-07-28	examples : fix whitespace	Georgi Gerganov
2023-07-28	examples : server chat mode with llama2 (#2400)	nhamanasu
2023-07-28	readme : fix the description of the Tail free sampling (TFS) method (#2431)	Weird Constructor
2023-07-28	llama : use n_embd_gqa instead of n_embd to handle llama-2 70B (#2433)	Rand Xie
2023-07-28	Obtaining LLaMA 2 instructions (#2308)	niansa/tuxifan
2023-07-27	convert.py : Update to support 70B HF format model files (#2427)	mj-shifu
2023-07-27	metal : disable graph concurrency optimization due to bug (#2413)	Georgi Gerganov
2023-07-26	ggml : fix assert in ggml_set_unary_op (#2410)	slaren
2023-07-26	make : build with -Wmissing-prototypes (#2394)	Cebtenzzre
2023-07-26	ggml : allocate graphs in a context (#2392)	slaren
2023-07-25	Add LLAMA_DEFAULT_RMS_EPS so we can change the default (#2384)	Kawrakow
2023-07-25	ggml : fix ggml_flash_attn to use op_params (#2387)	slaren
2023-07-25	convert.py : support bpe tokenizer (#2228)	ldwang
2023-07-25	ggml : relax contiguous constraints in activation function (#2371)	Jiahao Li
2023-07-25	ggml : improve graph build time via hash table lookup (#2329)	slaren
2023-07-25	build : fix line breaking error in build-info.sh (#2349)	Hesen Peng
2023-07-25	main : add `--in-prefix-bos` to prefix BOS to user inputs; keep EOS (#2304)	Xiao-Yong Jin
2023-07-25	ci : add non-AVX scalar build/test (#2356)	Eve
2023-07-25	k_quants : add AVX support to dot functions with QK_K as 64 (#2339)	katsu560
2023-07-25	metal : concurrently dispatch commands (#2358)	Shouzheng Liu
2023-07-25	Another speed gain for Q4_0 and Q4_1 on Metal (#2375)	Kawrakow
2023-07-25	Fix Q4_K and Q5_K for QK_K = 64 on CUDA (#2359)	Kawrakow
2023-07-25	server: add rms_norm_eps parameter (#2380)	slaren
2023-07-25	[Server] Escape HTML in webchat (#2368)	Henri Vasserman
2023-07-24	make rms_norm_eps a parameter (#2374)	slaren
2023-07-24	Chat UI extras (#2366)	Aarni Koskela
2023-07-24	ggml : sync (unary ops refactor, static-correctness) (#2370)	Georgi Gerganov
2023-07-24	Fix scalar version of Q5_K when QK_K = 64 (#2362)	Kawrakow
2023-07-23	llama : add grammar-based sampling (#1773)	Evan Jones
2023-07-24	Some more Q4_K and Q5_K speedup on CUDA (#2346)	Kawrakow
2023-07-23	Add gqa parameter support to the server (#2351)	IgnacioFDM
2023-07-23	Fix __dp4a documentation (#2348)	Johannes Gäßler
2023-07-23	common : n_threads == -1 uses std::thread::hardware_concurrency() (#2347)	wzy
2023-07-23	fix n_tasks (#2342)	slaren
2023-07-23	ggml: move op parameters from tensors to ggml_tensor::op_params (#2333)	slaren
2023-07-23	llama : grouped-query attention + LLaMAv2 70B support (#2276)	Georgi Gerganov
2023-07-23	llama : print help to stdout (#2338)	maddes8cht
2023-07-23	flake : support `nix build '.#opencl'` (#2337)	wzy
2023-07-23	llama : print max tensor size to stderr (#2336)	Christian Demsar
2023-07-23	make : fix CLBLAST compile support in FreeBSD (#2331)	Jose Maldonado
2023-07-23	examples : simplify vim plugin (#2327)	AustinMroz
2023-07-23	metal : support bcast add & dup & cont op (#2323)	Jiahao Li
2023-07-23	Speed up Q4_K (#2322)	Kawrakow
2023-07-22	CUDA: Fixed 7b q3_K_S with mul_mat_vec_q (#2313)	Johannes Gäßler
2023-07-22	llama : optimize memory buffers (#2325)	Georgi Gerganov
2023-07-22	Perplexity: Compute scores correlated to HellaSwag (#2312)	klosax
2023-07-22	examples : basic VIM plugin	whoreson
2023-07-22	ci : fix args	Georgi Gerganov
2023-07-22	ci : add 7B CUDA tests (#2319)	Georgi Gerganov