Mythos-class reasoning for autonomous knowledge work.
- Input
- 1,000 cr / M tokens
- Output
- 5,000 cr / M tokens
OpenGray routes your requests to Claude Fable 5 and GPT-5.6 Sol through a single OpenAI-compatible endpoint. Buy credits, generate a key, ship. No subscription, no seat licences, no minimum spend.
Credits never expire · Billed per token, to six decimal places · 100 credits = US$1.00
curl https://open-gray.com/api/v1/chat/completions \
-H "Authorization: Bearer sk-og-v1-..." \
-H "Content-Type: application/json" \
-d '{
"model": "claude-fable-5",
"messages": [
{ "role": "user", "content": "Summarise our Q3 revenue drivers." }
]
}'
from openai import OpenAI
client = OpenAI(
base_url="https://open-gray.com/api/v1",
api_key="sk-og-v1-...",
)
response = client.chat.completions.create(
model="claude-fable-5",
messages=[
{"role": "user", "content": "Summarise our Q3 revenue drivers."}
],
)
print(response.choices[0].message.content)
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://open-gray.com/api/v1",
apiKey: "sk-og-v1-...",
});
const response = await client.chat.completions.create({
model: "claude-fable-5",
messages: [
{ role: "user", content: "Summarise our Q3 revenue drivers." },
],
});
console.log(response.choices[0].message.content);
<?php
$payload = [
'model' => 'claude-fable-5',
'messages' => [
['role' => 'user', 'content' => 'Summarise our Q3 revenue drivers.'],
],
];
$ch = curl_init('https://open-gray.com/api/v1/chat/completions');
curl_setopt_array($ch, [
CURLOPT_POST => true,
CURLOPT_RETURNTRANSFER => true,
CURLOPT_HTTPHEADER => [
'Authorization: Bearer sk-og-v1-...',
'Content-Type: application/json',
],
CURLOPT_POSTFIELDS => json_encode($payload),
]);
$body = json_decode(curl_exec($ch), true);
echo $body['choices'][0]['message']['content'];
Catalogue
We carry the strongest reasoning model from each frontier lab instead of a directory of hundreds. Prices below are what you pay, in credits per million tokens.
Mythos-class reasoning for autonomous knowledge work.
Flagship GPT-5.6 model for coding and agentic work.
Platform
Everything you need to run production traffic, and nothing you don't.
Change two lines: the base URL and the key. Any OpenAI SDK, LangChain, or plain HTTP client works unchanged, streaming included.
Every request is metered against the generation's real token usage and settled to six decimals. No rounding up to whole cents.
Issue a key per environment, cap it in credits, and revoke it instantly. Keys are stored as SHA-256 hashes and shown once.
Per-request logs with tokens, latency, model and credits charged. Export-ready tables for finance and engineering.
Cached input tokens are billed at the cache rate, so repeat context in long agent loops stays inexpensive.
TLS everywhere, hashed credentials, row-level-secured storage and rate-limited authentication out of the box.
How it works
https://open-gray.com/api/v1 and call any model in the catalogue.Credits are pay-as-you-go and never expire. If OpenGray isn't a fit, stop sending requests — there is nothing to cancel.