Currently listed through these providers:
Model details
Cohere Command A (08/2025)
Command A 08/2025 marks Cohere's entry into reasoning-oriented models, purpose-built for enterprise customer service and automation. Marketed as a 111-billion-parameter dense model, it brings tool-use capabilities alongside broad multilingual coverage spanning 23 languages, including English, French, Spanish, Japanese, Arabic, and Hindi. This positions it well for global organizations needing to route and resolve customer interactions in a single deployment, without stitching together multiple language-specialized systems.
The model is engineered for practical agentic workflows rather than purely conversational chat. Its advertised strengths include document processing, scheduling, and data analysis, which map naturally onto the customer service and automation tasks Cohere highlights as its primary use case. The long context window supports ingesting lengthy tickets, policy documents, or knowledge-base articles in a single pass, while tool use enables the model to act on backend systems during a conversation. Practical fit is strongest for teams building retrieval-augmented assistants, automated support pipelines, and multilingual helpdesk tooling that needs coherent reasoning over large documents.
Quick Info
Powered by- Provider
- NanoGPT
- Model key
- command-a-reasoning-08-2025
- Release date
- Aug 21, 2025
- Last updated
- Aug 21, 2025
- Knowledge cutoff
- 2024-06-01
- Input modalities
- Output modalities
- Capabilities
Cost
- Input token cost
- $2.50
- Output token cost
- $10.00
Limits
- Input tokens
- 256,000 tokens
- Output tokens
- 8,192 tokens
- Context window
- 256,000 tokens
Transparent token rates
Compare Cohere Command A (08/2025) pricing
Rates are shown per one million tokens. Combined means one million input plus one million output tokens.
Latest news about Cohere Command A (08/2025)
No articles yet. Fetch the latest news to show it here.