infer0
OpenAI-compatible · /v1/chat/completions

Frontier AI APIs.
Up to 70% less.

One OpenAI-compatible endpoint for Claude, GPT, Grok, Gemini and the leading open models. Keep your SDK and change the base URL.

Get your API keySee pricing

$1 free credit · no card needed

  • Claude
  • ChatGPT
  • Grok
  • Gemini
  • Zhipu
  • Kimi
  • DeepSeek
  • MiniMax
model
import os
from openai import OpenAI
 
client = OpenAI(
base_url="https://api.infer0.ai/v1", # the only change
api_key=os.environ["INFER0_API_KEY"],
)
 
resp = client.chat.completions.create(
model="claude-sonnet-5",
messages=[{"role": "user", "content": "Summarize this incident report."}],
)
response
{"id": "chatcmpl-8f2c41…", "object": "chat.completion",
"model": "claude-sonnet-5",
"choices": [{"message": {"role": "assistant",
"content": "• Root cause: expired TLS cert on the edge proxy…"}}],
"usage": {"prompt_tokens": 1204, "completion_tokens": 386}}
✓ 200 OK · 1,204 in / 386 out tokens
cost $0.00439 list $0.00627 −30%

Example request, priced from the rate card below.

Models & pricing

Every major lab. One price list.

Each provider's list price, struck through, next to what you pay on infer0. Same models, same API.

USD per 1M tokens

ModelInput/MOutput/MCache write/MCache read/MLatencyThroughputUptime
Claude Opus 5.530% off
claude-opus-5-5
$4.00$2.80$20.00$14.00$5.00$3.50$0.20$0.14——
—
Claude Opus 5.5flex tier — Cheaper, lower reliability. The upstream may alter prompts; no injection guarantee.65% off
claude-opus-5-5:flex
$4.00$1.40$20.00$7.00$5.00$1.75$0.20$0.07——
—
Claude Sonnet 5.530% off
claude-sonnet-5-5
$2.00$1.40$10.00$7.00$2.50$1.75$0.20$0.14——
—
Claude Sonnet 5.5flex tier — Cheaper, lower reliability. The upstream may alter prompts; no injection guarantee.65% off
claude-sonnet-5-5:flex
$2.00$0.70$10.00$3.50$2.50$0.875$0.20$0.07——
—
Claude Opus 530% off
claude-opus-5
$5.00$3.50$25.00$17.50$6.25$4.375$0.50$0.35——
—
Claude Opus 5flex tier — Cheaper, lower reliability. The upstream may alter prompts; no injection guarantee.65% off
claude-opus-5:flex
$5.00$1.75$25.00$8.75$6.25$2.1875$0.50$0.175——
—
Claude Sonnet 530% off
claude-sonnet-5
$2.00$1.40$10.00$7.00$2.50$1.75$0.20$0.14——
—
Claude Sonnet 5flex tier — Cheaper, lower reliability. The upstream may alter prompts; no injection guarantee.65% off
claude-sonnet-5:flex
$2.00$0.70$10.00$3.50$2.50$0.875$0.20$0.07——
—
Claude Fable 5.130% off
claude-fable-5-1
$10.00$7.00$50.00$35.00$12.50$8.75$0.25$0.175——
—
Claude Fable 5.1flex tier — Cheaper, lower reliability. The upstream may alter prompts; no injection guarantee.65% off
claude-fable-5-1:flex
$10.00$3.50$50.00$17.50$12.50$4.375$0.25$0.0875——
—

List prices are each provider's published base rates. Where a model's prices change with the time of day or the length of a request, its row says how.

How it works

No migration. One line of config.

If your code already calls the OpenAI API, switching to infer0 is a config change, not a rewrite.

  1. 01

    Create an API key

    Sign up and start with $1 of free credit. No card needed.

    INFER0_API_KEY=sk-infer0-••••3f9a
  2. 02

    Change your base URL

    Keep your OpenAI SDK, prompts and tooling. One line of config is the whole migration.

    - api.openai.com/v1
    + api.infer0.ai/v1
  3. 03

    Call any frontier model

    Use the model names you already know, frontier or open. One key and one invoice for every lab.

    claude-sonnet-5gpt-5.6-terraglm-5.3deepseek-v4-pro

Stop paying list price for frontier models.

Every account starts with $1 of free credit.

Get your API keyLog in