 ymcki
|
bc7b1f8632
convert : fix Llama-3_1-Nemotron-51B rope settings (#11008)
|
hai 1 ano |
 Peter
|
6e1531aca5
common, examples, ggml : fix MSYS2 GCC compiler errors and warnings when building with LLAMA_CURL=ON and GGML_OPENCL=ON (#11013)
|
hai 1 ano |
 Jeff Bolz
|
716bd6dec3
vulkan: optimize mul_mat for small values of N (#10991)
|
hai 1 ano |
 ag2s20150909
|
c250ecb315
android : fix llama_batch free (#11014)
|
hai 1 ano |
|
Jeff Bolz
|
a813badbbd
vulkan: im2col and matmul optimizations for stable diffusion (#10942)
|
hai 1 ano |
|
Jeff Bolz
|
fdd2188912
vulkan: Use push constant offset to handle misaligned descriptors (#10987)
|
hai 1 ano |
 Isaac McFadyen
|
f865ea149d
server: added more docs for response_fields field (#10995)
|
hai 1 ano |
 Alexey Parfenov
|
16cdce7b68
server : fix token duplication when streaming with stop strings (#10997)
|
hai 1 ano |
 Eve
|
d79d8f39b4
vulkan: multi-row k quants (#10846)
|
hai 1 ano |
|
Peter
|
d283d02bf2
examples, ggml : fix GCC compiler warnings (#10983)
|
hai 1 ano |
 Reza Kakhki
|
9ba399dfa7
server : add support for "encoding_format": "base64" to the */embeddings endpoints (#10967)
|
hai 1 ano |
 Djip007
|
2cd43f4900
ggml : more perfo with llamafile tinyblas on x86_64 (#10714)
|
hai 1 ano |
 NeverLucky
|
09fe2e7613
server: allow filtering llama server response fields (#10940)
|
hai 1 ano |
 Georgi Gerganov
|
30caac3a68
llama : the WPM vocabs use the CLS token as BOS (#10930)
|
hai 1 ano |
 Diego Devesa
|
60cfa728e2
ggml : use wstring for backend search paths (#10960)
|
hai 1 ano |
|
Diego Devesa
|
3327bb0f8d
ggml : fix arm enabled features check (#10961)
|
hai 1 ano |
|
Diego Devesa
|
32d6ee6385
ggml : fix const usage in SSE path (#10962)
|
hai 1 ano |
 Xuan Son Nguyen
|
14b699ecde
server : fix missing model id in /model endpoint (#10957)
|
hai 1 ano |
|
Xuan Son Nguyen
|
485dc01214
server : add system_fingerprint to chat/completion (#10917)
|
hai 1 ano |
 Radoslav Gerganov
|
86bf31cfe6
rpc-server : add support for the SYCL backend (#10934)
|
hai 1 ano |
 Yun Dou
|
b92a14a841
llama : support InfiniAI Megrez 3b (#10893)
|
hai 1 ano |
|
ymcki
|
6f0c9e034b
llama : support for Llama-3_1-Nemotron-51B (#10669)
|
hai 1 ano |
 Eric Curtin
|
dab76c92cc
llama-run : include temperature option (#10899)
|
hai 1 ano |
 yuri@FreeBSD
|
7024d59e6a
ggml : fix run-time on FreeBSD in get_executable_path() (#10948)
|
hai 1 ano |
 Rudi Servo
|
7c0e285858
devops : add docker-multi-stage builds (#10832)
|
hai 1 ano |
 Billel Mokeddem
|
7ae33a616f
llama : add Falcon3 support (#10883)
|
hai 1 ano |
|
Jeff Bolz
|
ebdee9478c
vulkan: build fixes for 32b (#10927)
|
hai 1 ano |
|
Georgi Gerganov
|
5cd85b5e00
convert : add BertForMaskedLM (#10919)
|
hai 1 ano |
|
Jeff Bolz
|
a91a41364b
vulkan: optimize coopmat2 dequant functions (#10855)
|
hai 1 ano |
 Adrien Gallouët
|
e34c5af43f
ggml-cpu: replace NEON asm with intrinsics in ggml_gemv_q4_0_4x8_q8_0() (#10874)
|
hai 1 ano |
