ϟfluxgateAPI / 0.9

CHAT / INFERENCE

Stride 8B Instruct

Fast, general-purpose language inference for chat, structured extraction and tool calls.

POST/v1/chat/completions
CONTEXT / LIMIT32K tokens
OUTPUTJSON or SSE
MODEL IDstride-8b-instruct

Make a request

Set stream to true for a server-sent event stream. Use max_tokens to cap generated output.

curl https://YOUR_API_HOST/v1/chat/completions   -H 'Authorization: Bearer $FLUXGATE_KEY'   -H 'Content-Type: application/json'   -d '{"model":"stride-8b-instruct", "messages":[
    {"role":"user","content":"Explain token streaming."}
  ], "stream":true, "max_tokens":512}'

Request controls

ParameterTypeWhat it does
modelstringSelect a model ID from the catalog.
messagesarrayConversation turns with a role and text content.
max_tokensintegerUpper bound for tokens in the response.
streambooleanReturn incremental server-sent events.

Handle limits explicitly

A 429 response means the project rate or token budget has been reached. Respect the retry interval. A 401 response indicates a missing or invalid project key. Never expose production keys in browser code.