ngxson/llama.cpp

Pull Requests Commits

ngxson committed 354 days ago

450e47b2

add chat template test

ngxson committed 354 days ago

c6bd7a7a

fix compilation

ngxson committed 354 days ago

44f998af

server : clean up built-in template detection

ngxson committed 354 days ago

c5ac2b85

common, examples, ggml : fix MSYS2 GCC compiler errors and warnings when building with LLAMA_CURL=ON and GGML_OPENCL=ON (#11013)

peter277 committed 355 days ago

Verified 6e1531ac

vulkan: optimize mul_mat for small values of N (#10991)

jeffbolznv committed 355 days ago

Verified 716bd6de

android : fix llama_batch free (#11014)

ag2s20150909 committed 355 days ago

Verified c250ecb3

vulkan: im2col and matmul optimizations for stable diffusion (#10942)

jeffbolznv committed 356 days ago

Verified a813badb

vulkan: Use push constant offset to handle misaligned descriptors (#10987)

jeffbolznv committed 356 days ago

Verified fdd21889

server: added more docs for response_fields field (#10995)

isaac-mcfadyen committed 357 days ago

Verified f865ea14

server : fix token duplication when streaming with stop strings (#10997)

z80maniac committed 357 days ago

Verified 16cdce7b

vulkan: multi-row k quants (#10846)

netrunnereve committed 359 days ago

Verified d79d8f39

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

peter277 committed 359 days ago

Verified d283d02b

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

elk-cloner committed 361 days ago

Verified 9ba399df

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

Djip007 committed 361 days ago

Verified 2cd43f49

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

nvrxq committed 361 days ago

Verified 09fe2e76

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

ggerganov committed 361 days ago

Verified 30caac3a

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

slaren committed 362 days ago

Verified 60cfa728

ggml : fix arm enabled features check (#10961)

slaren committed 362 days ago

Verified 3327bb0f

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

slaren committed 362 days ago

Verified 32d6ee63

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

ngxson committed 362 days ago

Verified 14b699ec

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

ngxson committed 362 days ago

Verified 485dc012

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

rgerganov committed 362 days ago

Verified 86bf31cf

llama : support InfiniAI Megrez 3b (#10893)

dixyes committed 363 days ago

Verified b92a14a8

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

ymcki committed 363 days ago

Verified 6f0c9e03

llama-run : include temperature option (#10899)

ericcurtin committed 363 days ago

Verified dab76c92

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

yurivict committed 363 days ago

Verified 7024d59e

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

rudiservo committed 363 days ago

Verified 7c0e2858

llama : add Falcon3 support (#10883)

mokeddembillel committed 363 days ago

Verified 7ae33a61

vulkan: build fixes for 32b (#10927)

jeffbolznv committed 363 days ago

Verified ebdee947

Older