SporeLabs API Reference

SporeLabs is an OpenAI-compatible inference endpoint. Change your base_url and api_key, keep your existing client.

Base URL:  https://sporelabs.dev/v1
Auth:      Authorization: Bearer <api-key>

If you already have code that calls OpenAI, swap two lines and everything keeps working:

# Before
client = OpenAI(api_key="sk-...")

# After
client = OpenAI(
    base_url="https://sporelabs.dev/v1",
    api_key="sl-xxx...",
)

Request/response shapes, auth headers, SSE streaming format, tool calls, JSON mode, and error envelopes all follow the OpenAI API contract. What is different:

1. Model IDs use OpenRouter's provider/model format (or omit model entirely — see Model IDs).
2. Base URL is https://sporelabs.dev/v1 (not https://api.openai.com/v1).
3. Payment is a prepaid dollar wallet debited at real upstream spend (pass-through, no markup), not postpaid billing.
4. Extra endpoints like /v1/improve/settings are SporeLabs-only additions beyond the OpenAI spec.

Quick Start (copy-paste)

Python (openai SDK)

from openai import OpenAI

client = OpenAI(
    base_url="https://sporelabs.dev/v1",
    api_key="sl-xxx...",
)

response = client.chat.completions.create(
    model="openai/gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

Omit model (or send "auto") to let SporeLabs pick: traffic starts on Kimi K3 (moonshotai/kimi-k3) and only moves when a cheaper model is proven equal on your traffic.

curl

curl https://sporelabs.dev/v1/chat/completions \
  -H "Authorization: Bearer sl-xxx..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-4o",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

JavaScript (openai npm)

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://sporelabs.dev/v1",
  apiKey: "sl-xxx...",
});

const response = await client.chat.completions.create({
  model: "openai/gpt-4o",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

Anthropic SDK

The Anthropic surface (/v1/messages) is served from the same base URL — use x-api-key or Bearer with your SporeLabs key:

import anthropic

client = anthropic.Anthropic(
    base_url="https://sporelabs.dev/v1",
    api_key="sl-xxx...",
)

message = client.messages.create(
    model="anthropic/claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello!"}],
)
print(message.content[0].text)

Streaming (SSE)

stream = client.chat.completions.create(
    model="openai/gpt-4o",
    messages=[{"role": "user", "content": "Count to 5"}],
    stream=True,
)
for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="")

OpenAI-Compatible Endpoints

These endpoints match the OpenAI API shape so any OpenAI SDK works with a base URL swap.

POST /v1/chat/completions

Create a chat completion.

Request body (OpenAI ChatCompletionRequest):

Response (OpenAI ChatCompletion):

{
  "id": "chatcmpl-xxx",
  "object": "chat.completion",
  "created": 1698850000,
  "model": "openai/gpt-4o",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello! How can I help you?"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 7,
    "total_tokens": 19
  }
}

When stream=true, each chunk follows the OpenAI SSE format:

data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","created":1698850000,"model":"openai/gpt-4o","choices":[{"index":0,"delta":{"role":"assistant","content":""},"finish_reason":null}]}

data: {"choices":[{"index":0,"delta":{"content":"Hello"},"finish_reason":null}]}

data: {"choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}

data: [DONE]

GET /v1/models

List available models: a minimal set — auto, the Kimi incumbent, and owned adapters — not the full OpenRouter index.

Response:

{
  "object": "list",
  "data": [
    {
      "id": "openai/gpt-4o",
      "object": "model",
      "created": 1698850000,
      "owned_by": "openai"
    }
  ]
}

GET /v1/models/{model_id}

Retrieve a specific model by its OpenRouter slug.

SporeLabs-Specific Endpoints

These are not part of the OpenAI standard but use the same base URL and bearer token. Full details in the Jobs API doc.

Authentication

All endpoints use standard bearer authentication:

Authorization: Bearer <api-key>

Get your API key from the SporeLabs portal dashboard.

Errors

Errors follow the OpenAI error envelope:

{
  "error": {
    "message": "Insufficient funds",
    "type": "insufficient_quota",
    "param": null,
    "code": "insufficient_quota"
  }
}

Common error types:

Model IDs

Omit model (or send auto) and SporeLabs routes for you: Auto starts on Kimi K3 (moonshotai/kimi-k3) and moves a job to a cheaper model only after proving equality on your traffic. Name a model and that exact model serves the call.

Named models use OpenRouter slugs. Common ones:

moonshotai/kimi-k3
openai/gpt-4o
openai/gpt-4o-mini
openai/o3-mini
anthropic/claude-sonnet-4-20250514
anthropic/claude-3.5-haiku-20241022
google/gemini-2.5-pro-preview-03-25
meta-llama/llama-3.3-70b-instruct

Get the current list from GET /v1/models.

Limits

• Context window: Depends on model (8K-200K tokens).
• Max response tokens: Depends on model (4K-128K).
• Streaming: Supported for all chat models.
• Rate limits: Contact support if you need higher limits for your workload.