Benchmarked models

Open frontier reasoning and orchestration MoE on a hybrid Transformer-Mamba architecture.

120B-parameter open hybrid MoE that activates 12B per token for complex multi-agent work.

Small, compute-efficient MoE for building specialized agentic AI systems.

Open multimodal 30B-A3B model designed as a perception and context sub-agent.

12B open multimodal reasoning model for video understanding and document intelligence.

9B dense LLM trained from scratch as a unified model for reasoning and non-reasoning tasks.

Open-weight 21B MoE released under Apache 2.0 with 3.6B active parameters per forward pass.

Cohere’s first agentic coding model, optimized for code generation and repo-level work.

Google DeepMind instruction-tuned MoE with near-31B quality while activating only 3.8B per token.

Google DeepMind 30.7B dense multimodal model with 256K context and configurable thinking mode.

MoE from inclusionAI for responsive agents, instruction following, and multi-turn conversation.

Latest Poolside coding agent model, scoring 70.2% on Terminal-Bench 2.1.

33B-A3B coding agent model from Poolside, a step forward from Laguna XS.2.