Skip to content
GitLab
Explore
Sign in
Tags
Tags give the ability to mark specific points in history as being important
This project is mirrored from
https://github.com/ggerganov/llama.cpp
. Pull mirroring updated
Sep 19, 2024
.
b3609
2f3c1466
·
llava: Add ACC OP for GPU acceleration to the Vulkan backend in the LLAVA CLIP model. (#8984)
·
Aug 20, 2024
b3608
50addec9
·
[SYCL] fallback mmvq (#9088)
·
Aug 20, 2024
b3607
4f8d19ff
·
[SYCL] Fix SYCL `im2col` and `convert` Overflow with Large Dims (#9052)
·
Aug 20, 2024
b3606
90db8146
·
tests : add missing comma in grammar integration tests (#9099)
·
Aug 20, 2024
b3604
1b6ff90f
·
rpc : print error message when failed to connect endpoint (#9042)
·
Aug 19, 2024
b3603
18eaf29f
·
rpc : prevent crashes on invalid input (#9040)
·
Aug 19, 2024
b3600
2fb92678
·
Fix incorrect use of ctx_split for bias tensors (#9063)
·
Aug 17, 2024
b3599
8b3befc0
·
server : refactor middleware and /health endpoint (#9056)
·
Aug 16, 2024
b3598
d565bb2f
·
llava : support MiniCPM-V-2.6 (#8967)
·
Aug 16, 2024
b3593
fb487bb5
·
common : add support for cpu_get_num_physical_cores() on Windows (#8771)
·
Aug 16, 2024
b3592
2a24c8ca
·
Add Nemotron/Minitron GGUF Conversion & Inference Support (#8922)
·
Aug 16, 2024
b3591
e3f6fd56
·
ggml : dynamic ggml_sched_max_splits based on graph_size (#9047)
·
Aug 16, 2024
b3590
4b9afbbe
·
retrieval : fix memory leak in retrieval query handling (#8955)
·
Aug 15, 2024
b3589
37501d9c
·
server : fix duplicated n_predict key in the generation_settings (#8994)
·
Aug 15, 2024
b3588
4af8420a
·
common : remove duplicate function llama_should_add_bos_token (#8778)
·
Aug 15, 2024
b3587
6bda7ce6
·
llama : add pre-tokenizer regexes for BLOOM and gpt3-finnish (#8850)
·
Aug 15, 2024
b3585
234b3067
·
server : init stop and error fields of the result struct (#9026)
·
Aug 15, 2024
b3584
5fd89a70
·
Vulkan Optimizations and Fixes (#8959)
·
Aug 14, 2024
b3583
98a532d4
·
server : fix segfault on long system prompt (#8987)
·
Aug 14, 2024
b3582
43bdd3ce
·
cmake : remove unused option GGML_CURL (#9011)
·
Aug 14, 2024
Prev
1
…
3
4
5
6
7
8
9
10
11
…
123
Next