 Yoshi Suhara
|
2a24c8caa6
Add Nemotron/Minitron GGUF Conversion & Inference Support (#8922)
|
hai 1 ano |
 Nico Bosshard
|
e3f6fd56b1
ggml : dynamic ggml_sched_max_splits based on graph_size (#9047)
|
hai 1 ano |
 gtygo
|
4b9afbbe90
retrieval : fix memory leak in retrieval query handling (#8955)
|
hai 1 ano |
 Riceball LEE
|
37501d9c79
server : fix duplicated n_predict key in the generation_settings (#8994)
|
hai 1 ano |
 Zhenwei Jin
|
4af8420afb
common : remove duplicate function llama_should_add_bos_token (#8778)
|
hai 1 ano |
 Esko Toivonen
|
6bda7ce6c3
llama : add pre-tokenizer regexes for BLOOM and gpt3-finnish (#8850)
|
hai 1 ano |
 Georgi Gerganov
|
d5492f0525
ci : disable bench workflow (#9010)
|
hai 1 ano |
 Jiří Podivín
|
234b30676a
server : init stop and error fields of the result struct (#9026)
|
hai 1 ano |
 0cc4m
|
5fd89a70ea
Vulkan Optimizations and Fixes (#8959)
|
hai 1 ano |
 compilade
|
98a532d474
server : fix segfault on long system prompt (#8987)
|
hai 1 ano |
|
Georgi Gerganov
|
43bdd3ce18
cmake : remove unused option GGML_CURL (#9011)
|
hai 1 ano |
 Daniel Bevenius
|
06943a69f6
ggml : move rope type enum to ggml.h (#8949)
|
hai 1 ano |
 Xuan Son Nguyen
|
828d6ff7d7
export-lora : throw error if lora is quantized (#9002)
|
hai 1 ano |
 Diogo Teles Sant'Anna
|
fc4ca27b25
ci : fix github workflow vulnerable to script injection (#9008)
|
hai 1 ano |
 Radoslav Gerganov
|
1f67436c5e
ci : enable RPC in all of the released builds (#9006)
|
hai 1 ano |
|
Nico Bosshard
|
0fd93cdef5
llama : model-based max number of graph nodes calculation (#8970)
|
hai 1 ano |
 Frank Mai
|
84eb2f4fad
docs: introduce gpustack and gguf-parser (#8873)
|
hai 1 ano |
 DavidKorczynski
|
1262e7ed13
grammar-parser : fix possible null-deref (#9004)
|
hai 1 ano |
|
DavidKorczynski
|
df5478fbea
ggml: fix div-by-zero (#9003)
|
hai 1 ano |
 Liu Jia
|
2589292cde
Fix a spelling mistake (#9001)
|
hai 1 ano |
|
Georgi Gerganov
|
d3ae0ee8d7
py : fix requirements check '==' -> '~=' (#8982)
|
hai 1 ano |
|
Georgi Gerganov
|
5ef07e25ac
server : handle models with missing EOS token (#8997)
|
hai 1 ano |
|
compilade
|
4134999e01
gguf-py : Numpy dequantization for most types (#8939)
|
hai 1 ano |
|
Georgi Gerganov
|
8cd1bcfd3f
flake.lock: Update (#8979)
|
hai 1 ano |
 Neo Zhang
|
a21c6fd450
update guide (#8909)
|
hai 1 ano |
 fairydreaming
|
33309f661a
llama : check all graph nodes when searching for result_embd_pooled (#8956)
|
hai 1 ano |
 Markus Tavenrath
|
7c5bfd57f8
Optimize Vulkan backend for better CPU performance and less GPU synchronization overhead. (#8943)
|
hai 1 ano |
 slaren
|
6e02327e8b
metal : fix uninitialized abort_callback (#8968)
|
hai 1 ano |
|
Xuan Son Nguyen
|
7eb23840ed
llama : default n_swa for phi-3 (#8931)
|
hai 1 ano |
|
fairydreaming
|
7c3f55c100
Add support for encoder-only T5 models (#8900)
|
hai 1 ano |
