API Documentation

The API is OpenAI-compatible. Point any OpenAI SDK at this host and go.

Base URL

https://inferenceyield.ai/v1

Also available at https://api.inferenceyield.ai/v1.

Chat completions (streaming)

curl https://inferenceyield.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen/qwen3.8-27b",
    "stream": true,
    "stream_options": {"include_usage": true},
    "messages": [{"role": "user", "content": "Write a haiku about GPUs."}]
  }'

OpenAI SDK (Python)

from openai import OpenAI
client = OpenAI(base_url="https://inferenceyield.ai/v1", api_key="none-during-pilot")
r = client.chat.completions.create(
    model="qwen/qwen3.8-27b",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(r.choices[0].message.content)

Features

• Streaming (SSE) with usage in the final chunk when stream_options.include_usage is set.
• Tool calling — pass tools; parallel calls supported.
• Structured outputs / JSON mode via response_format.
• Prefix caching — repeated context is billed at the cached-input rate; see usage.prompt_tokens_details.cached_tokens on every response.
• Usage accounting on stream and non-stream responses alike.

Model metadata

GET /v1/models returns pricing, context length, max output tokens, quantization, supported features, and datacenter location for every listed model.

Limits & keys

The pilot endpoint is open with fair-use limits. For an API key, sustained throughput, or ZDR contract terms: [email protected].