deepseek-ai/DeepSeek-V4-Pro-Base
1.6T • Updated • 23.1k • 284
Base Models
Some of the relevant (May/2026) language base models on Hugging Face (meaning, LLMs trained on pure text, without Instruct/SFT/RLHF/RL tuni etc)
-
-
deepseek-ai/DeepSeek-V4-Flash-Base
292B • Updated • 48.2k • 229 -
deepseek-ai/DeepSeek-V3-Base
685B • Updated • 6.55k • 1.7k -
deepseek-ai/DeepSeek-V3.1-Base
Text Generation • 685B • Updated • 6.85k • 1.01k -
moonshotai/Kimi-K2-Base
Text Generation • 1T • Updated • 43.1k • 303 -
meta-llama/Llama-3.1-405B
Text Generation • 406B • Updated • 192k • • 976 -
arcee-ai/Trinity-Large-Base
Text Generation • 399B • Updated • 405 • 57 -
google/gemma-4-26B-A4B
Image-Text-to-Text • 27B • Updated • 134k • 287 -
google/gemma-4-31B
Image-Text-to-Text • 33B • Updated • 535k • 395 -
Qwen/Qwen3.5-0.8B-Base
Image-Text-to-Text • 0.9B • Updated • 187k • 78 -
Qwen/Qwen3.5-35B-A3B-Base
Image-Text-to-Text • 36B • Updated • 97.6k • 131 -
Qwen/Qwen3.5-9B-Base
Image-Text-to-Text • 10B • Updated • 131k • 79 -
talkie-lm/talkie-1930-13b-base
Updated • 88 -
nvidia/Nemotron-Labs-Diffusion-14B-Base
Text Generation • 14B • Updated • 4.05k • 6 -
baidu/ERNIE-4.5-300B-A47B-Base-PT
Text Generation • 299B • Updated • 103 • • 42 -
XiaomiMiMo/MiMo-V2.5-Pro-Base
Text Generation • 1T • Updated • 1.6k • 37 -
stepfun-ai/Step-3.5-Flash-Base
Text Generation • 198B • Updated • 151 • 84 -
zai-org/GLM-4.5-Base
Text Generation • 358B • Updated • 345 • 60 -
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-Base-BF16
Text Generation • 124B • Updated • 28.6k • 30 -
ibm-granite/granite-4.1-30b-base
Text Generation • 29B • Updated • 3.37k • 27 -
arcee-ai/Trinity-Nano-Base
Text Generation • 6B • Updated • 1.93k • 27 -
arcee-ai/Trinity-Mini-Base
Text Generation • 26B • Updated • 1.99k • 21