SpiderGate API Reference

SpiderGate is SpiderIQ's OpenAI-compatible LLM gateway — one endpoint in front of 100+ models across 18 providers. Send a request to a task alias like spideriq/fast and SpiderGate routes it to the best available model, falls back automatically on failure, caps your spend, and hands back the real per-call cost.

• Base URL — https://spideriq.ai/api/gate/v1

• Protocol — OpenAI Chat Completions, drop-in. Point any OpenAI SDK at the base URL above.

• Auth — a Bearer API key (see Authentication).

• Current version — 2.70.0

Why a gateway

You send model: "spideriq/coding" instead of a specific provider model. SpiderGate:

• Routes to the best model in that alias's chain,

• Falls back to the next provider if one is down or rate-limited,

• Caps cost per request (spidergate_options.max_cost_usd),

• Caches identical prompts, and

• Reports the real dollar cost of every call in spidergate_metadata.cost_usd.

One key replaces juggling OpenRouter + Groq + Mistral + Cerebras credentials.

Your first request

curl https://spideriq.ai/api/gate/v1/chat/completions \
  -H "Authorization: Bearer $SPIDERGATE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "spideriq/fast",
    "messages": [
      { "role": "user", "content": "In one sentence, what is an LLM gateway?" }
    ],
    "max_tokens": 60
  }'

Response (real call, served by nvidia_nim/meta/llama-3.1-8b-instruct in ~327 ms):

{
  "id": "chatcmpl-721027d3a2444b09a1769d21",
  "object": "chat.completion",
  "created": 1784494387,
  "model": "nvidia_nim/meta/llama-3.1-8b-instruct",
  "choices": [
    {
      "index": 0,
      "finish_reason": "stop",
      "message": {
        "role": "assistant",
        "content": "An LLM (Large Language Model) gateway is a software interface that enables other applications or services to access and utilize the capabilities of a large language model, such as natural language processing and generation, in a secure and scalable manner."
      }
    }
  ],
  "usage": { "prompt_tokens": 52, "completion_tokens": 47, "total_tokens": 99 }
}

You never named a provider — spideriq/fast picked the model.

Where to next