ggerganov/llama.cpp

Pull Requests Commits

profiler: initial support for profiling graph ops

max-krasnyansky committed 149 days ago

a362c74a

vulkan: multi-row k quants (#10846)

netrunnereve committed 149 days ago

Verified d79d8f39

examples, ggml : fix GCC compiler warnings (#10983)

peter277 committed 150 days ago

Verified d283d02b

server : add support for "encoding_format": "base64" to the */embeddings endpoints (#10967)

elk-cloner committed 151 days ago

Verified 9ba399df

ggml : more perfo with llamafile tinyblas on x86_64 (#10714)

Djip007 committed 151 days ago

Verified 2cd43f49

server: allow filtering llama server response fields (#10940)

nvrxq committed 151 days ago

Verified 09fe2e76

llama : the WPM vocabs use the CLS token as BOS (#10930)

ggerganov committed 152 days ago

Verified 30caac3a

ggml : use wstring for backend search paths (#10960)

slaren committed 152 days ago

Verified 60cfa728

ggml : fix arm enabled features check (#10961)

slaren committed 152 days ago

Verified 3327bb0f

ggml : fix const usage in SSE path (#10962)

slaren committed 152 days ago

Verified 32d6ee63

server : fix missing model id in /model endpoint (#10957)

ngxson committed 153 days ago

Verified 14b699ec

server : add system_fingerprint to chat/completion (#10917)

ngxson committed 153 days ago

Verified 485dc012

rpc-server : add support for the SYCL backend (#10934)

rgerganov committed 153 days ago

Verified 86bf31cf

llama : support InfiniAI Megrez 3b (#10893)

dixyes committed 153 days ago

Verified b92a14a8

llama : support for Llama-3_1-Nemotron-51B (#10669)

ymcki committed 153 days ago

Verified 6f0c9e03

llama-run : include temperature option (#10899)

ericcurtin committed 153 days ago

Verified dab76c92

ggml : fix run-time on FreeBSD in get_executable_path() (#10948)

yurivict committed 153 days ago

Verified 7024d59e

devops : add docker-multi-stage builds (#10832)

rudiservo committed 153 days ago

Verified 7c0e2858

llama : add Falcon3 support (#10883)

mokeddembillel committed 153 days ago

Verified 7ae33a61

vulkan: build fixes for 32b (#10927)

jeffbolznv committed 154 days ago

Verified ebdee947

convert : add BertForMaskedLM (#10919)

ggerganov committed 155 days ago

Verified 5cd85b5e

vulkan: optimize coopmat2 dequant functions (#10855)

jeffbolznv committed 155 days ago

Verified a91a4136

ggml-cpu: replace NEON asm with intrinsics in ggml_gemv_q4_0_4x8_q8_0() (#10874)

angt committed 155 days ago

Verified e34c5af4

SYCL: Migrate away from deprecated ggml_tensor->backend (#10840)

qnixsynapse committed 155 days ago

Verified eb5c3dc6

server : (UI) fix copy to clipboard function (#10916)

ngxson committed 156 days ago

Verified 0ca416c9

ggml : add test for SVE and disable when it fails (#10906)

slaren committed 156 days ago

Verified 21ae3b9b

convert : fix RWKV v6 model conversion (#10913)

MollySophia committed 156 days ago

Verified 0a11f8b7

clip : disable GPU support (#10896)

ggerganov committed 156 days ago

Verified d408bb92

llama : minor grammar refactor (#10897)

ggerganov committed 156 days ago

Verified 5cab3e4a

tts : small QoL for easy model fetch (#10903)

ggerganov committed 156 days ago

Verified 36319dec

Older