New Chat
Ollama Cloud
Qwen 3.5
Alibaba — general purpose
DeepSeek V4 Pro
Flagship reasoning
DeepSeek V4 Flash
Fast variant
DeepSeek V3.2
Balanced MoE
Kimi K2.6
Moonshot — long context
GLM 5.1
Zhipu — latest
GLM 5
Zhipu — strong reasoning
GLM 4.7
Zhipu — efficient
MiniMax M3
Latest
MiniMax M2.7
Large variant
MiniMax M2.5
Balanced
MiniMax M2.1
Efficient
Gemini 3 Flash
Google — fast multimodal
Gemma 4
Google — open weights
Nemotron 3 Ultra
NVIDIA — largest
Nemotron 3 Super
NVIDIA — strong performance
Nemotron 3 Nano
NVIDIA — efficient
Devstral Small 2
Mistral/MPI — coding
RNJ-1
Reasoning model
Qwen3 Coder Next
Alibaba — code specialist
AI
Press Enter to send, Shift+Enter for new line