Models

All models available on HuggingChat

meta-llama/Llama-3.3-70B-Instruct logo Active
meta-llama/Llama-3.3-70B-Instruct Ideal for everyday use. A fast and extremely capable model matching closed source models' capabilities. Now with the latest Llama 3.3 weights!
Qwen/Qwen3-235B-A22B logo
Qwen/Qwen3-235B-A22B Qwen's flagship model featuring optional reasoning. Exceptional performance with benchmarks rivaling R1 and o1.
Qwen/Qwen2.5-72B-Instruct logo
Qwen/Qwen2.5-72B-Instruct The latest Qwen open model with improved role-playing, long text generation and structured data understanding.
CohereForAI/c4ai-command-r-plus-08-2024 logo
CohereForAI/c4ai-command-r-plus-08-2024 Cohere's largest language model, optimized for conversational interaction and tool use. Now with the 2024 update!
deepseek-ai/DeepSeek-R1-Distill-Qwen-32B logo
deepseek-ai/DeepSeek-R1-Distill-Qwen-32B The first reasoning model from DeepSeek, distilled into a 32B dense model. Outperforms o1-mini on multiple benchmarks.
nvidia/Llama-3.1-Nemotron-70B-Instruct-HF logo
nvidia/Llama-3.1-Nemotron-70B-Instruct-HF Nvidia's latest Llama fine-tune, topping alignment benchmarks and optimized for instruction following.
Qwen/QwQ-32B logo
Qwen/QwQ-32B QwQ is the latest reasoning model released by the Qwen team, approaching the capabilities of R1 in benchmarks.
google/gemma-3-27b-it logo
google/gemma-3-27b-it Google's latest open model with great multilingual performance, supports image inputs natively.
mistralai/Mistral-Small-3.1-24B-Instruct-2503 logo
mistralai/Mistral-Small-3.1-24B-Instruct-2503 A small model with good capabilities in language understanding and commonsense reasoning.
Qwen/Qwen2.5-VL-32B-Instruct logo
Qwen/Qwen2.5-VL-32B-Instruct The latest multimodal model from Qwen! Supports image inputs natively.
microsoft/Phi-4 logo
microsoft/Phi-4 One of the best small models, super fast for simple tasks.
NousResearch/Hermes-3-Llama-3.1-8B logo
NousResearch/Hermes-3-Llama-3.1-8B Nous Research's latest Hermes 3 release in 8B size. Follows instruction closely.