 Jeff Bolz
|
5b3466bedf
vulkan: Handle GPUs with less shared memory (#10468)
|
il y a 1 an |
|
Jeff Bolz
|
249a7902ec
vulkan: further optimize q5_k mul_mat_vec (#10479)
|
il y a 1 an |
|
Jeff Bolz
|
71a64989a5
vulkan: skip integer div/mod in get_offsets for batch_idx==0 (#10506)
|
il y a 1 an |
|
Jeff Bolz
|
4a57d362e1
vulkan: optimize Q2_K and Q3_K mul_mat_vec (#10459)
|
il y a 1 an |
 Diego Devesa
|
c9b00a70b0
ci : fix cuda releases (#10532)
|
il y a 1 an |
 Shane A
|
de5097351c
Add OLMo 2 model in docs (#10530)
|
il y a 1 an |
|
Diego Devesa
|
5a349f2809
ci : remove nix workflows (#10526)
|
il y a 1 an |
|
Diego Devesa
|
30ec398321
llama : disable warnings for 3rd party sha1 dependency (#10527)
|
il y a 1 an |
 Tristan Druyen
|
be0e350c8b
Fix HIP flag inconsistency & build docs (#10524)
|
il y a 1 an |
 R0CKSTAR
|
249cd93da3
mtgpu: Add MUSA_DOCKER_ARCH in Dockerfiles && update cmake and make (#10516)
|
il y a 1 an |
|
Jeff Bolz
|
904109ed0d
vulkan: fix group_norm (#10496)
|
il y a 1 an |
 Xuan Son Nguyen
|
45abe0f74e
server : replace behave with pytest (#10416)
|
il y a 1 an |
 Neo Zhang Jianyu
|
0bbd2262a3
restore the condistion to build & update pacakge when merge (#10507)
|
il y a 1 an |
 Georgi Gerganov
|
ab96610b1e
cmake : enable warnings in llama (#10474)
|
il y a 1 an |
|
Diego Devesa
|
7db3846a94
ci : publish the docker images created during scheduled runs (#10515)
|
il y a 1 an |
|
Diego Devesa
|
c6807b3f28
ci : add ubuntu cuda build, build with one arch on windows (#10456)
|
il y a 1 an |
 Charles Xu
|
25669aa92c
ggml-cpu: cmake add arm64 cpu feature check for macos (#10487)
|
il y a 1 an |
|
Georgi Gerganov
|
84e1c33cde
server : fix parallel speculative decoding (#10513)
|
il y a 1 an |
|
Georgi Gerganov
|
811872a59d
speculative : simplify the implementation (#10504)
|
il y a 1 an |
 Shanshan Shen
|
9a4b79bcfa
CANN: Improve the Inferencing Performance for Ascend NPU Device (#10454)
|
il y a 1 an |
 Chenguang Li
|
7066b4cce2
CANN: RoPE and CANCAT operator optimization (#10488)
|
il y a 1 an |
 Junil Kim
|
0eb4e12bee
vulkan: Fix a vulkan-shaders-gen arugment parsing error (#10484)
|
il y a 1 an |
 Eric Curtin
|
0cc63754b8
Introduce llama-run (#10291)
|
il y a 1 an |
|
Diego Devesa
|
50d5cecbda
ci : build docker images only once daily (#10503)
|
il y a 1 an |
|
Georgi Gerganov
|
9fd8c2687f
server : add more information about error (#10455)
|
il y a 1 an |
|
Georgi Gerganov
|
47f931c8f9
server : enable cache_prompt by default (#10501)
|
il y a 1 an |
|
Georgi Gerganov
|
106964e3d2
metal : enable mat-vec kernels for bs <= 4 (#10491)
|
il y a 1 an |
|
Shane A
|
80acb7b430
Rename Olmo1124 to Olmo2 (#10500)
|
il y a 1 an |
|
Diego Devesa
|
10bce0450f
llama : accept a list of devices to use to offload a model (#10497)
|
il y a 1 an |
 Johannes Gäßler
|
1f922254f0
Github: update issue templates [no ci] (#10489)
|
il y a 1 an |
