Nemotron 3.5 Super VL vs Router.com (Ramp Router)

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.

Router.com (Ramp Router)

Detailed Markdown docs with typed error codes, per-model capability data and routing an agent can set per request. No OpenAPI document, status page, changelog or published rate limits were found. Signup needs a US business email in a browser, and Ramp records prompts and outputs for a year by default unless switched off.

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

Router.com (Ramp Router)

1. Switch off content recording in Settings before sending private data. It is on by default and keeps prompts and outputs for a year
2. Read model ids from GET /v1/models at runtime. Ids are account-specific, an unknown one is a 404, and Gemini ids need your own Google key
3. Use POST /v1/responses with a models list for fallbacks. Router stops falling back once a stream has started
4. Set fallback_timeout and provider_timeout, or allow_flex_tier: false, when latency matters. Flex routing is on by default
5. Treat both 401 invalid_api_key and 401 api_key_deactivated as a possibly locked key, and stop retrying on 402