Explore Mixes
Crosshatch Mixes combine the strengths of multiple AI models to deliver optimal performance for specific tasks. Each mix is carefully curated and automatically updated to ensure you always have access to the best AI capabilities without manual selection.
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.
A custom-built Mixture-of-Agents (MoA) synthesis mix optimized for challenging coding tasks. This mix leverages multiple 'proposer' models, including Claude 3.5 Sonnet and GPT-4 Turbo, with an 'aggregation' model that synthesizes their outputs. In benchmarks, it demonstrated 28% better performance compared to Claude 3.5 Sonnet alone, particularly excelling at complex programming challenges.
A fast variant of the Coding Mixture of Agents (MoA) that uses GPT-4o-mini to classify task difficulty. Simple tasks are quickly processed by GPT-4o, while complex ones utilize the full MoA. With 68% classification accuracy and a bias towards the MoA when uncertain, this mix optimizes for speed without compromising on quality for challenging problems.
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.
Get Your Own Custom Mixes
Submit your use case or a general description of what you need for a custom tuned solution.