cturan/llama.cpp

作者	SHA1 備註	提交日期
Kawrakow	66d575c45c llama : add Q3_K_XS (#5060)	2 年之前
Georgi Gerganov	44a1a4a41a backend : add eval callback (#4935)	2 年之前
David Friehs	4483396751 llama : apply classifier-free guidance to logits directly (#4951)	2 年之前
Kawrakow	147b17ac94 2-bit quantizations (#4897)	2 年之前
David Friehs	df845cc982 llama : minimize size used for state save/load (#4820)	2 年之前
slaren	e7e4df031b llama : ggml-backend integration (#4766)	2 年之前
Kawrakow	469e75d0a3 llama : restore intended k-quants mixes for MoE models (#4872)	2 年之前
Kawrakow	49662cbed3 ggml : SOTA 2-bit quants (add IQ2_XS) (#4856)	2 年之前
Kawrakow	dd5ae06405 SOTA 2-bit quants (#4773)	2 年之前
Georgi Gerganov	52531fdff8 main : add self-extend support (#4815)	2 年之前
Georgi Gerganov	b0034d93ce examples : add passkey test (#3856)	2 年之前
Marcus Dunn	0040d42eeb llama : replace all API facing `int`'s with `int32_t` (#4577)	2 年之前
crasm	c7e9701f86 llama : add ability to cancel model loading (#4462)	2 年之前
Marcus Dunn	31f27758fa llama : allow getting n_batch from llama_context in c api (#4540)	2 年之前
slaren	c6c4fc081c lora : add support for non-llama models (#3333)	2 年之前
crasm	6391817cd1 llama : document logits_all deprecation (#4418)	2 年之前
Georgi Gerganov	bcc0eb4591 llama : per-layer KV cache + quantum K cache (#4309)	2 年之前
Kerfuffle	5aa365d88f llama : allow overriding GGUF metadata when loading model (#4092)	2 年之前
crasm	3014b5415d Update docs for yarn_ext_factor <0.0 as unspecified instead of NaN (#4189)	2 年之前
Georgi Gerganov	6b0a7420d0 llama : KV cache view API + better KV cache management (#4170)	2 年之前
slaren	e85bb1a8e7 llama : add functions to get the model's metadata (#4013)	2 年之前
Kerfuffle	91f6499393 Respect tokenizer.ggml.add_bos_token value when tokenizing (#4040)	2 年之前
Georgi Gerganov	05816027d6 common : YAYF (yet another YARN fix) (#3925)	2 年之前
cebtenzzre	898aeca90a llama : implement YaRN RoPE scaling (#2268)	2 年之前
kalomaze	238657db23 samplers : Min-P sampler implementation [alternative to Top P/Top K] (#3841)	2 年之前
Kerfuffle	6e08281e58 Extend llama_kv_cache_seq_rm to allow matching any sequence (#3843)	2 年之前
Georgi Gerganov	d69d777c02 ggml : quantization refactoring (#3833)	2 年之前
Georgi Gerganov	ee1a0ec9cb llama : add option for greedy sampling with probs (#3813)	2 年之前
Georgi Gerganov	2f9ec7e271 cuda : improve text-generation and batched decoding performance (#3776)	2 年之前
Marcus Dunn	5be6c803fa llama : remove token functions with `context` args in favor of `model` (#3720)	2 年之前

更新的提交更舊的提交

提交歷史 查找

提交歷史