Novita AI vs Beam 1.0

Novita AI

Per-token prices for about 140 language models are public, and each key can carry an expiry, a model allowlist and a source IP allowlist. No status page, SLA or security.txt was found, the per-model rate limit figures are drawn by script, and both vendor SDK repositories are archived while the docs still point to them.

Beam 1.0

The Reflection API is OpenAI-compatible, with an OpenAPI document, strict structured outputs and tool calls, and error codes that say when to retry, and the weights are Apache-2.0. It is a free public beta with unpublished daily limits and no per-token price, reasoning can't be turned off, and three API incidents were posted between 14 and 16 October 2026.

Novita AI

1. Set the OpenAI client base URL to https://api.novita.ai/openai and send the key as Authorization: Bearer. Keys start with sk_
2. Do not close a connection to stop generation. A request that reached the model is billed in full under status 499, so cap output with max_tokens
3. On 429, check whether the code is RATE_LIMIT_EXCEEDED or TOKEN_LIMIT_EXCEEDED and back off exponentially. No Retry-After header is documented
4. Read the changelog for retirement notices before pinning a model id. One retirement in October 2026 came with 11 days' notice
5. Ask the account owner for a key with an expiry, a model access policy and an IP allowlist. Keys are created and deleted only in the console

Beam 1.0

1. Send reasoning_effort low or medium for short tasks. The default is max, and none or minimal return 400 unsupported_value
2. Return every assistant message with its reasoning_content unchanged. If a turn comes back without it, start a new conversation, since later requests fail with missing_required_parameter
3. Don't rely on stop or seed. Both are accepted and ignored, so truncate output in your own code
4. Honour x-should-retry: false on a 429. It means a daily limit is spent and Retry-After counts to 00:00 UTC
5. Set max_completion_tokens yourself on long prompts. The default budget isn't reduced to fit the prompt and includes reasoning tokens