curl – fremai docs

curl

Raw HTTP is the lowest-common-denominator client — useful for scripts, CI, and debugging. All endpoints take a Bearer key and return standard OpenAI-compatible JSON.

Set your key once:

export FREMAI_API_KEY="sk-fremai-…"

Chat completion

curl https://api.fremai.eu/v1/chat/completions \
  -H "Authorization: Bearer $FREMAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-5.2",
    "messages": [
      {"role": "system", "content": "You are a concise assistant."},
      {"role": "user", "content": "What is EU data sovereignty in one sentence?"}
    ],
    "temperature": 0.3
  }'

Streaming

Add "stream": true. The response is a Server-Sent Events stream of chat.completion.chunk objects, terminated by data: [DONE]:

curl -N https://api.fremai.eu/v1/chat/completions \
  -H "Authorization: Bearer $FREMAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-5.2",
    "messages": [{"role": "user", "content": "Stream a short greeting."}],
    "stream": true
  }'

Embeddings

curl https://api.fremai.eu/v1/embeddings \
  -H "Authorization: Bearer $FREMAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "<embedding-model-id>",
    "input": ["EU-sovereign inference", "OpenAI-compatible API"]
  }'

List models

curl https://api.fremai.eu/v1/models \
  -H "Authorization: Bearer $FREMAI_API_KEY"

Returns the catalog available to your key. Each id is a value you can pass as model.

Tips

Use -N (--no-buffer) when streaming so curl prints chunks as they arrive.
On a 429, respect the Retry-After response header before retrying. See error codes.
The /v1 prefix is the advertised form; the API also answers the bare-root path (e.g. /chat/completions) for clients configured without it.

OpenAI SDK LangChain