Cohere
Command A+
Cohere's first Mixture-of-Experts model, combining vision input, agentic tool use, reasoning and translation in one enterprise deployment.
Our editorial assessment, based on the public sources listed below. Last reviewed .
- Overall rank
- #73
- Tier
- B
- Score
- 68.3
- Main category
- Chat
Strengths
- Single model covers vision, agentic tool use, reasoning and translation
- Fits on 1x B200 or 2x H100 GPUs for on-prem enterprise deployment
- Notable latency and throughput gains over Command A Reasoning
Considerations
- 128K context is smaller than the 1M-token windows of frontier rivals
- Enterprise-oriented positioning means less consumer tooling around it
Price
Open weights (Apache 2.0); API $0.30 input / $1.50 output per 1M tokens via OpenRouter; also on Azure AI Foundry
Shown for reference — price is not part of the score.
Score breakdown
- Performance
- 66.0
- 55% of score
- Reputation
- 68.0
- 35% of score
- Recency
- 82.0
- 10% of score
The overall score combines these sub-scores using the weights shown. How we score →
Verification status
Each field carries its own re-verification date. Amber means older than 90 days — treat those figures as stale until re-checked.
- Pricing
- Benchmark
- Reputation
- Feature list
Checked against the sources listed below.
Sources
- Command A+ — Artificial Analysis (opens in a new tab)Accessed:
- Command A+ Apache 2.0 open model — VentureBeat (opens in a new tab)Accessed:
- Command A+ 218B sparse MoE — MarkTechPost (opens in a new tab)Accessed:
- Cohere releases Command A+ (Business Wire) (opens in a new tab)Accessed:
- An Overview of Cohere's Models (opens in a new tab)Accessed:
Show 1 more source
Similar models
- Gemini 3 FlashGoogle#69 · Tier A · Score 71.1Compare
- Mistral LargeMistral AI#71 · Tier B · Score 69.2Compare
- Gemini 2.5 FlashGoogle#75 · Tier B · Score 68.2Compare
- GPT-5 miniOpenAI#77 · Tier B · Score 64.9Compare