 Jared Van Bortel
|
2f567611c0
llama-model : support Qwen2 embedding models and pooling_mode_lasttoken (#13245)
|
il y a 9 mois |
|
Jared Van Bortel
|
7d2123484e
convert : use correct context length for nomic-embed-text-v2 (#13216)
|
il y a 9 mois |
 Xuan-Son Nguyen
|
074e42ab31
convert : converting mmproj for Qwen2/2.5VL from convert_hf_to_gguf (#13209)
|
il y a 9 mois |
|
Xuan-Son Nguyen
|
dcf886007d
convert : explicitly disable trust_remote_code for AutoConfig (#13246)
|
il y a 9 mois |
|
Xuan-Son Nguyen
|
8936784f7a
mtmd : add **vision** support for Mistral Small 3.1 (#13231)
|
il y a 9 mois |
|
Xuan-Son Nguyen
|
3e168bede4
convert : improve model arch handling (#13122)
|
il y a 9 mois |
|
Xuan-Son Nguyen
|
07c2e2f76c
convert : correct typo image_mean --> image_std (#13208)
|
il y a 9 mois |
 AT
|
5f5e39e1ba
model : Nomic Embed Text V2 with Mixture-of-Experts (MoE) architecture (#12466)
|
il y a 9 mois |
 matteo
|
ced44be342
llama-chat : fix wrong template in GLM4-0414 (#13140)
|
il y a 9 mois |
 HimariO
|
ca2bb89eac
clip : Add Qwen2.5VL support (#12402)
|
il y a 9 mois |
|
Xuan-Son Nguyen
|
ecda2ec4b3
mtmd : Support Pixtral 12B (#13065)
|
il y a 9 mois |
 piDack
|
eb1776b15a
convert : Append mult-eos,half-rope,bos to GLM4-0414 and Z (#13021)
|
il y a 9 mois |
|
Xuan-Son Nguyen
|
dc39a5e7a8
mtmd : support SmolVLM (version 1 and 2) (#13050)
|
il y a 9 mois |
|
Xuan-Son Nguyen
|
2016f07bd1
convert : experimental support for `--mmproj` flag (#13023)
|
il y a 9 mois |
 Juk Armstrong
|
daa422881a
llama : DeepSeek V2/V3 MLA implementation (#12801)
|
il y a 9 mois |
 Yuxuan Zhang
|
06bb53ad9b
llama-model : add Glm4Model implementation for GLM-4-0414 (#12867)
|
il y a 9 mois |
 Daniel Han
|
ec6c09d0fa
convert : Llama4 RoPE fix (#12889)
|
il y a 10 mois |
|
Xuan-Son Nguyen
|
5b1f13cb64
convert : proper tensor name mapping for llama4 (#12870)
|
il y a 10 mois |
|
Xuan-Son Nguyen
|
64eda5deb9
convert : ability to lazy-load safetensors remotely without downloading to disk (#12820)
|
il y a 10 mois |
 Bo Zheng
|
d3bd7193ba
llama : Support Qwen3 and Qwen3MoE (#12828)
|
il y a 10 mois |
|
Xuan-Son Nguyen
|
1466621e73
llama : Support llama 4 text-only (#12791)
|
il y a 10 mois |
 Sigbjørn Skjæret
|
5936a616e4
convert : BailingMoE : fix qkv split when head_dim is 0 (#12687)
|
il y a 10 mois |
|
Sigbjørn Skjæret
|
35782aeedb
convert : BailingMoE : avoid setting rope_dim to 0 (#12678)
|
il y a 10 mois |
|
Sigbjørn Skjæret
|
403fbacbbc
convert : Qwerky : use lora_rank_tokenshift and lora_rank_decay if present (#12667)
|
il y a 10 mois |
|
Sigbjørn Skjæret
|
2c3f8b850a
llama : support BailingMoE (Ling) (#12634)
|
il y a 10 mois |
 Juyoung Suk
|
b3de7cac73
llama : add Trillion 7B model support (#12556)
|
il y a 10 mois |
 Si1w
|
f125b8dccf
llama : add PLM GGUF Conversion & Inference Support (#12457)
|
il y a 10 mois |
 Csaba Kecskemeti
|
d5c6309d91
convert : Support Qwen2_5_VLForConditionalGeneration (#12595)
|
il y a 10 mois |
 Georgi Gerganov
|
df4d20cd53
convert : fix squeeze for ssm_conv tensors (#12573)
|
il y a 10 mois |
|
Sigbjørn Skjæret
|
53af4dba42
convert: fix Mistral3/Gemma3 model hparams init (#12571)
|
il y a 10 mois |
