ggerganov/llama.cpp

Pull Requests Commits

Update ggml/src/ggml.c

max-krasnyansky committed 246 days ago

Verified 6b0248c2

ggml : fix n_threads_cur initialization with one thread

slaren committed 246 days ago

f9196c91

llama : use reserve/emplace_back in sampler_sample (#9534)

danbev committed 246 days ago

Verified 6443ddd9

server : match OAI structured output response (#9527)

VJHack committed 247 days ago

Verified 8a308354

server : fix OpenSSL build (remove obsolete `LOG_INFO`) (#9529)

EZForever committed 247 days ago

Verified f799155a

[SYCL]set context default value to avoid memory issue, update guide (#9476)

NeoZhangJianyu committed 247 days ago

Verified faf67b3d

llama-bench: correct argument parsing error message (#9524)

Xarbirus committed 247 days ago

Verified 7be099fa

arg : add env variable for parallel (#9513)

bertwagner committed 247 days ago

Verified 8b836ae7

llama : fix n_vocab init for 'no_vocab' case (#9511)

Xarbirus committed 248 days ago

Verified 8344ef58

threadpool : skip polling for unused threads (#9461)

max-krasnyansky committed 248 days ago

Verified 02266138

unicode : add <algorithm> (#9508)

ykhrustalev committed 248 days ago

Verified 503147a9

llama : support IBM Granite architecture (#9412)

gabe-l-hart committed 248 days ago

Verified 0d2ec438

llama : add llama_n_head() (#9512)

Xarbirus committed 248 days ago

Verified 37f3a381

ggml : move common CPU backend impl to new header (#9509)

slaren committed 248 days ago

Verified 23e0d70b

llama : rename n_embed to n_embd in rwkv6_time_mix (#9504)

danbev committed 249 days ago

Verified acb2c32c

ggml : link MATH_LIBRARY not by its full path (#9339)

Xarbirus committed 249 days ago

Verified a6a3a5c5

convert : identify missing model files (#9397)

compilade committed 249 days ago

Verified d54c21df

cmake : do not hide GGML options + rename option (#9465)

ggerganov committed 249 days ago

Verified 19514d63

ggml : IQ4_NL sgemm + Q4_0 AVX optimization (#9422)

netrunnereve committed 249 days ago

Verified 5c3d0f18

llama : support OLMoE (#9462)

2015aroras committed 249 days ago

Verified 0aadac10

llama : support MiniCPM3 (#9322)

CarryFun committed 249 days ago

Verified 95ca8516

main : option to disable context shift (#9484)

VJHack committed 249 days ago

Verified 441b72b9

metal : handle zero-sized allocs (#9466)

ggerganov committed 249 days ago

Verified c4965a64

flake.lock: Update (#9488)

ggerganov committed 249 days ago

Verified 90a2fff0

common : reimplement logging (#9418)

ggerganov committed 249 days ago

Verified 6262d13e

gguf-split : add basic checks (#9499)

slaren committed 249 days ago

Verified e6deac31

cmake : correct order of sycl flags (#9497)

Xarbirus committed 249 days ago

Verified 6988da94

py : add "LLaMAForCausalLM" conversion support (#9485)

csabakecskemeti committed 250 days ago

Verified 3c7989fd

readme : update tools list (#9475)

OLSecret committed 250 days ago

Verified d6b37c88

cmake : try to fix sycl+intel build (#9487)

Xarbirus committed 250 days ago

Verified 7596487b

Older