deepgrove/maple-preview
Text Generation • 20B • Updated • 1.73k • 433
New Architectures, or such
because Qwen/GLM/Gemma/etc gets boring. A lot of these also seem to be unsupported in mainline llama.cpp ... so I'm waiting for that.
-
-
thinkingmachines/Inkling-Small
Image-Text-to-Text • 266B • Updated • 1.2M • • 454 -
upstage/Solar-Open2-250B
Text Generation • 250B • Updated • 8.17k • 760 -
tencent/Hy3
Text Generation • 299B • Updated • 1.14M • • 991 -
ai9stars/G9v3-39A5B
Text Generation • 39B • Updated • 917 • 106 -
sarvamai/sarvam-105b
Text Generation • 106B • Updated • 5.77k • 326
tencent/Hy4-preview
Text Generation • 780B • Updated • 11.5k • • 553Note FINALLY a GLM5-size (but even larger) competitor. dear gods, it's been so long since there's been a 700-800B class model with more than 40B params.
-
Motif-Technologies/Motif-3
Text Generation • 315B • Updated • 2.37k • 154 -
XHToken/Spark-X2.5-4B
Text Generation • 4B • Updated • 28.1k • 1.44k -
IFM/K2-Horizon-MoVA-36B-A4B
Text Generation • 37B • Updated • 15.1k • • 432 -
Nanbeige/Nanbeige4.2-3B
Text Generation • 4B • Updated • 108k • 859 -
XiaomiMiMo/MiMo-V2.6-Flash-RL
Text Generation • 311B • Updated • 76.9k • • 593 -
Aleph-Alpha/Kolibri-1
Text Generation • 78B • Updated • 14.5k • 1.01k -
RecursiveSI/flint
Text Generation • 40B • Updated • 47.8k • 1.39k -
reflection/Beam-1.0-501B-A23B
Text Generation • 501B • Updated • 12.9k • • 318 -
inclusionAI/Ling-3.1-flash
Text Generation • 572B • Updated • 7.85k • • 167 -
stepfun-ai/Step-5-Preview
604B • Updated • 7.73k • 139 -
XHToken/Spark-X2.6-4B
Text Generation • 4B • Updated • 26k • 1.04k -
baidu/ERNIE-5.5-Flash-Base
Text Generation • 286B • Updated • 2.11k • 249 -
taobao-mnn/MNN-Pocket-0.3B-GGUF
Text Generation • 0.4B • Updated • 7.85k • 15
