A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA. The model is multilingual, supporting English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese,...
Rank
#54
Up 18% week over week
Tokens
139.5B
2026-08-14
Requests
60.0M
59,955,968
Tool calls
-
Input / M
$0.02
Output / M
$0.03
Cache / M
Free
Context
131K
131,072 tokens