CN8 API Documentation

Chat completion with a model catalog, per-model pricing, streaming, and structured outputs.

Send messages and get back data.message, data.model, data.finishreason, data.tokenusage. The cost object reports what was deducted from your balance. Aliases (default, gpt-4.1-mini, gpt4, gpt4-mini, gpt-4.1-nano, gpt-4.1) resolve to canonical model_ids.

• GET core-llm-models: models with post-markup pricing, default_model, aliases, tiers
• POST core-llm-completions: messages, model, temperature, maxtokens, stream, jsonschema → data.message, data.model, data.finishreason, data.tokenusage, optional data.parsed
• GET core-llm-models-comparison: side-by-side table with inputprice, outputprice, costpermessage