API Reference

Chat Completions

OpenAI-compatible chat endpoint — drop-in for any OpenAI SDK or tool.

Endpoint

POST https://api.prismma.ai/v1/chat/completions — the same request and response shape as OpenAI's Chat Completions API. Any tool or SDK that speaks that API works with Prismma without modification, other than the base URL and key.

Auth: Authorization: Bearer sk-eu-... — see Authentication for the full header table.

Request

shell
curl https://api.prismma.ai/v1/chat/completions \
  -H "Authorization: Bearer $PRISMMA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-opus-4-7",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

model is any id from Guides → Available Models that your key's provider group can serve. messages follows the standard OpenAI chat message array. Set stream: true for server-sent events — see Guides → Streaming.

Migrating existing code

If you already call OpenAI directly, switching to Prismma is a one-line change: swap the base URL and key, keep everything else.

shell
# Before: api.openai.com/v1
# After: change the host only — everything else stays

curl https://api.prismma.ai/v1/chat/completions \
  -H "Authorization: Bearer $PRISMMA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"claude-opus-4-7","messages":[{"role":"user","content":"Hello"}]}'
Model ids differ from OpenAI's — pass a Prismma model id (e.g. claude-sonnet-4-6, gpt-5.5), not an api.openai.com model name. See Available Models.