OpenAI's open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.

A mixture-of-experts model offering superior performance across various tasks, by Mistral team.

Deploy and run your own DeepSeek R1 model for advanced language processing and generation capabilities.

Latest version of DeepSeek's general-purpose language model.

High-performance language model with strong reasoning capabilities.

Meta's state-of-the-art large language model optimized for dialogue use cases.

A fine-tuned version of Mixtral offering enhanced instruction following capabilities.

Multimodal version of LLaMA 3.2 with vision capabilities.

Code-specialized version of Qwen 2.5 optimized for programming tasks.

General purpose language model with strong performance.

NVIDIA's high-performance language model for general tasks.

Second generation of the efficient small language model.

Advanced code generation model trained on a diverse set of programming languages.

Advanced code generation model with strong performance on programming tasks.

Latest version of the Qwen model series with enhanced performance.

Latest version of Microsoft's Phi model series with improved capabilities.

Code-specialized language model optimized for programming tasks.

Specialized code generation model with multiple size options.

NVIDIA-optimized version of Mistral for enhanced performance.

Advanced language model with strong multilingual capabilities.

Efficient small language model for resource-constrained environments.

Microsoft's efficient language model known for strong reasoning capabilities.

Meta's code-specialized language model optimized for programming tasks.

Code-specialized version of Google's Gemma model for programming tasks.

Second generation of Google's Gemma model with improved capabilities.

Advanced code generation model with multilingual programming capabilities.

Code-specialized version of Qwen optimized for programming tasks.

Versatile language model with strong multilingual capabilities.

Versatile language model with strong general capabilities.

A versatile language model with multiple size options for different use cases.

Google's efficient and powerful language model for various tasks.

Advanced language model focused on instruction following and reasoning.

Fine-tuned language model optimized for chat and instruction following.

Multimodal model capable of understanding and reasoning about images and text.

Open-source chat model optimized for dialogue interactions.

Compact and efficient language model suitable for resource-constrained environments.

Second generation of the Nous Hermes language model.

Original StarCoder model for code generation and understanding.

Fine-tuned language model known for high-quality chat interactions.

Uncensored version of Wizard-Vicuna for open-ended interactions.

Fine-tuned version of Mistral optimized with the OpenOrca dataset.

Chinese-optimized version of LLaMA 2 for enhanced Chinese language capabilities.