Cohere
Cohere's open-weight multimodal mixture-of-experts model for vision, reasoning, translation, and agentic tasks, with 218B total and 25B active parameters.
Running this yourself: likely needs a high-memory cloud gpu.
24.7
Quality Score
---
Arena ELO
218B
Parameters
128K
Context
Use this section to answer one simple question first: how much outside evidence do we have that this model performs well? Structured benchmark scores appear first, then official provider evidence, then live arena signal.
This model belongs to a benchmark-expected category, but stable public benchmark evidence has not been ingested yet.
A full normalized benchmark row is not available for this model yet.
This usually means public benchmark sources have not published stable data for this release yet.
Sign in to join the discussion
0
Downloads
0
Likes
May 2026
Released