# Cohere Chat API (Command models) (slim)

> Cohere's Chat API generates text with the Command family of models over REST (v2 chat), with tool use, structured outputs, citations and streaming. SDKs cover Python, TypeScript, Java and Go.

- Full: https://www.anchorterminal.com/tools/cohere-chat.md (~8,650 tokens) · this version ~1,780 tokens · JSON https://www.anchorterminal.com/tools/cohere-chat.json · canonical https://www.anchorterminal.com/tools/cohere-chat
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-10

**B · 64.7/100 · rank #342 of 950 · #11 in Model APIs & inference · not agent-ready · confidence medium**


Assessment: A public OpenAPI file, Markdown twins of every docs page and SDKs in four languages make the Chat API easy for an agent to read. Production keys do not cover the four newest Command models, whose limits are set by sales, and the SaaS agreement lets Cohere share API data with third parties.

## Facts

- Kind: Model API · vendor: Cohere · category: Model APIs & inference · legal entity: Cohere Inc. · provenance 83/100
- Endpoint: `https://api.cohere.com/v2/chat` (HTTP)
- Auth: API key · pricing: Freemium · x402: no · licence: Proprietary service under Cohere's Commercial SaaS Agreement. The SDKs are MIT
- Probe metrics: not measured yet (probes haven't run)
- Endpoint: `POST /v2/chat` on api.cohere.com per the OpenAPI file (the guide's examples use api.cohere.ai). `model` and `messages` required, 21 request fields, SSE streaming with `stream`
- Live Command models: `command-a-plus-05-2026` (128k context, 64k output, text and images), `command-a-03-2025` (256k, 8k), `command-a-reasoning-08-2025` (256k, 32k), `command-a-vision-07-2025` (128k, 8k), `command-a-translate-08-2025` (8k, 8k), `command-r-plus-08-2024`, `command-r-08-2024` and `command-r7b-12-2024` (128k, 4k)
- Rate limits: Trial 20 requests a minute per chat model and 1,000 calls a month. Production 500 a minute for Command A, R+, R, R7B and North Mini Code. Contact sales for Command A+, Reasoning, Translate and Vision
- Free tier: Trial key at sign-up, no card, not for production or commercial use
- Trains on API data: Paying customers can opt out under Data Controls. Trial use falls under the website terms and privacy policy
- Data retention: 30 days for logged prompts and generations unless flagged. Zero data retention by approval
- Data location: Google Cloud US-Central per the trust centre. Nine sub-processors listed, all in the US
- Structured output and tools: `response_format` with `json_object` or a JSON Schema, `tools` with `tool_choice` REQUIRED or NONE, `strict_tools` (beta), citations from `documents`, `thinking` with a token budget
- Batch: `/v2/batches` creates, lists, reads and cancels batches from an uploaded dataset. No batch price found
- Errors: 400, 401, 402, 404, 429, 499, 500 and others, each with sample messages. Bodies carry `message` and `id`, no error code. No Retry-After documented
- SDKs: Python `cohere` 7.1.0, TypeScript `cohere-ai` 8.1.0, Java 1.11.0 and Go v2.19.0, all tagged 26 August 2026, MIT
- Status: status.cohere.com on incident.io, a component per model, three incidents between 23 July and 1 September 2026
- Certifications: SOC 2 Type II, ISO 27001, ISO 42001 and UK Cyber Essentials per the trust centre, with a bug bounty under a responsible disclosure policy
- Other deployments: Model Vault (Command A at $40 to $48 an instance-hour, by waitlist), private deployments, Amazon Bedrock and SageMaker, Azure AI Foundry and Oracle OCI
- Models ($/1M in/out): `command-r-plus-08-2024` $2.50/$10
- 2026-04-04 Shutdown: `embed-english-v2.0`, `embed-multilingual-v2.0`, `c4ai-aya-expanse-8b` and `c4ai-aya-vision-8b` retired
- 2025-09-15 Notice: `command`, `command-light`, `command-r-03-2024`, `command-r-plus-04-2024`, `/v1/generate`, `/v1/summarize`, `/v1/classify`, connectors and fine-tuning deprecated

- Scores: Reliability 64, Performance pending, Schema & documentation 84, Agent ergonomics 72, Security & auth 57, Payments & pricing 30, Task success pending, Maintenance & community 72, Transparency & trust 74 · total over the 7 assessed categories
- Why: Reliability, Hosted lines. · Schema & documentation, Model reading. · Agent ergonomics, Model reading. · Security & auth, Model reading. · Payments & pricing, No x402, MPP or L402 in the docs, the OpenAPI file or the pricing page (0). · Maintenance & community, Model reading. · Transparency & trust, Closed service under a published SaaS agreement, MIT SDKs, and the docs say Command A+ weights are on Hugging Face under Apache 2.0, which w…
- Sources: 22, open questions: 11, both in the full twin
- Capabilities: inference.llm, inference.open-weights
- JSON: https://www.anchorterminal.com/api/v1/tools/cohere-chat.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/cohere-chat.svg` or a link to https://www.anchorterminal.com/tools/cohere-chat from a page on cohere.com or one of its subdomains, or the README of github.com/cohere-ai/cohere-python, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Call `POST https://api.cohere.com/v2/chat` with `Authorization: bearer <key>`. `model` and `messages` are the only required fields
2. Use `command-a-03-2025`, `command-r-08-2024`, `command-r-plus-08-2024` or `command-r7b-12-2024` for production traffic. Newer variants stay at trial limits on a production key
3. Keep a trial key under 20 chat requests a minute and 1,000 calls a month, and do not use it for commercial work
4. Set `response_format` with a `json_schema` for structured output, and tell the model to produce JSON when using `json_object` without a schema
5. Ask the account Owner to turn off training under Data Controls in the dashboard before sending confidential prompts

## Connect

```bash
pip install cohere   # or: npm i cohere-ai
```

```bash
curl --request POST \
  --url https://api.cohere.ai/v2/chat \
  --header 'accept: application/json' \
  --header 'content-type: application/json' \
  --header "Authorization: bearer $CO_API_KEY" \
  --data '{"model":"command-a-plus-05-2026","messages":[{"role":"user","content":"Hello"}]}'
```

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| GroqCloud | BB | 75.6 | inference.llm, inference.open-weights | https://www.anchorterminal.com/tools/groq.min.md |
| Mistral AI API | BB | 71.2 | inference.llm, inference.open-weights | https://www.anchorterminal.com/tools/mistral-api.min.md |
| Cloudflare Workers AI | B | 68.3 | inference.llm, inference.open-weights | https://www.anchorterminal.com/tools/cloudflare-workers-ai.min.md |
| SambaCloud | B | 66.4 | inference.llm, inference.open-weights | https://www.anchorterminal.com/tools/sambanova.min.md |
| DeepInfra | B | 63 | inference.llm, inference.open-weights | https://www.anchorterminal.com/tools/deepinfra.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)