
Command A
Cohere's 111B-parameter enterprise LLM built for RAG, tool use and agentic work on just two GPUs.
About
Command A is Cohere's flagship enterprise model, released March 2025, with 111B parameters and a 256K-token context window. It is tuned for retrieval-augmented generation, multilingual work (23 languages) and agentic tool use, and runs on two A100/H100 GPUs where comparable models need far more. The research weights on Hugging Face are CC-BY-NC, so commercial use goes through Cohere's API; its main weakness is that open-weights enterprise models like this still trail the very largest proprietary systems on raw frontier tasks.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (verified tier, metrics-v1). Evidence: March 2025 launch with follow-up Reasoning, Vision and Translate variants through Aug 2025; Real production traffic visible via OpenRouter; integrated into Cohere's North agent platform; Weights limited to CC-BY-NC research release; commercial path is Cohere API; Official changelog, HF model card and pricing cross-checked across three independent sources. License: Open weights under CC-BY-NC 4.0 (research, non-commercial) plus Cohere Labs Acceptable Use Policy; commercial use via Cohere API..
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 7.0/10
- Maintenance · recency, releases, issue load
- 8.0/10
- Openness · license + docs
- 5.0/10
- Manual review · sanity read + red-flag check
- 8.0/10
- · March 2025 launch with follow-up Reasoning, Vision and Translate variants through Aug 2025
- · Real production traffic visible via OpenRouter; integrated into Cohere's North agent platform
- · Weights limited to CC-BY-NC research release; commercial path is Cohere API
- · Official changelog, HF model card and pricing cross-checked across three independent sources