Explore Mixes

A specialized open-source mix optimized for engaging, multi-turn conversations and roleplay. Primarily powered by Hermes 3, a fine-tuned Llama 3.1 405B model, it excels in long-term context retention and adaptive responses.

A mix that picks the highest-ranked open-source model for Chinese prompts, based on LMSYS Chinese Leaderboard.

The best non-commercial coding models according to the LMSys leaderboard. This mix will use the top coding models ranked in the LMSys Chatbot Arena for the Coding category without any proprietary models.

A mix that picks the highest-ranked model for French prompts, based on LMSYS French Leaderboard.

A mix that picks the highest-ranked model for German prompts, based on LMSYS German Leaderboard.

A mix that picks the highest-ranked model for Japanese prompts, based on LMSYS Japanese Leaderboard.

A mix that picks the highest-ranked model for Korean prompts, based on LMSYS Korean Leaderboard.

Stay effortlessly current with the best general-purpose open-source AI models. This mix automatically utilizes the top-ranked open source models from the LMSys Chatbot Arena's overall leaderboard, ensuring you always have access to the most capable AI for a wide range of tasks.

A mix that picks the highest-ranked model for Russian prompts, based on LMSYS Russian Leaderboard.

A mix that picks the highest-ranked model for Spanish prompts, based on LMSYS Spanish Leaderboard.

A Mixture-of-Agents (MoA) synthesis mix optimized for fast and balanced chat use cases. This mix leverages multiple 'proposer' models, including Llama 3.1 70B on Groq and DeepSeek v2.5, with an 'aggregation' model that synthesizes their outputs.

An open-source Mixture-of-Agents (MoA) synthesis mix optimized for coding tasks. This mix leverages two 'proposer' models, DeepSeek-2.5 and Mistral Large 2, with Llama 3.1 405B as the 'aggregation' model that synthesizes their outputs. This mix offers a powerful, cost-effective solution for complex programming challenges.