Cohere Chat API (Command models) vs Nemotron 3.5 Super VL

Cohere Chat API (Command models)

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.

Nemotron 3.5 Super VL

NVIDIA's free endpoint takes OpenAI-format requests with typed tool calls, JSON Schema output and an OpenAPI 3.1 file, and the weights run anywhere under a permissive licence. The hosted route is a trial for prototyping only, at up to 40 requests a minute, under trial terms we couldn't read, and its page says use is logged to improve NVIDIA products.

Cohere Chat API (Command models)

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

Nemotron 3.5 Super VL

1. Set chat_template_kwargs to {"enable_thinking": false} for short tasks. Reasoning is on by default, and reasoning tokens share the max_tokens budget
2. Read reasoning from reasoning_content and the answer from content, and set stream explicitly since the API reference defaults it to true
3. Don't send confidential or personal data to the hosted trial. The model page asks callers not to and says use is logged
4. Back off and retry on throttling. Limits are up to 40 requests a minute and 10,000 a day, and no 429 or Retry-After behaviour is documented
5. For production, use a partner endpoint or serve the weights. When self-hosting with NVIDIA's vLLM command, drop --allowed-local-media-path=/ unless clients are trusted