Tokenly
Developers
The integration target is an OpenAI-compatible surface, so existing SDKs keep working and model IDs stay exactly as their developers publish them.
Integration overview
Requests carry a model ID from this catalogue. Switching models means changing one string. Base URL, keys and routing will be issued with the Phase 2 portal.
Example calls
PlannedIllustrative only. These snippets are not connected to a live service and will return nothing until Phase 2 ships.
curl https://api.tokenly.example/v1/chat/completions \
-H "Authorization: Bearer $TOKENLY_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "Qwen/Qwen2.5-72B-Instruct",
"messages": [{ "role": "user", "content": "Summarise this contract clause." }]
}'import OpenAI from 'openai';
const client = new OpenAI({
baseURL: 'https://api.tokenly.example/v1',
apiKey: process.env.TOKENLY_API_KEY,
});
const response = await client.chat.completions.create({
model: 'Qwen/Qwen2.5-72B-Instruct',
messages: [{ role: 'user', content: 'Summarise this contract clause.' }],
});from openai import OpenAI
client = OpenAI(
base_url="https://api.tokenly.example/v1",
api_key=os.environ["TOKENLY_API_KEY"],
)
response = client.chat.completions.create(
model="Qwen/Qwen2.5-72B-Instruct",
messages=[{"role": "user", "content": "Summarise this contract clause."}],
)What is not available yet
No API keys, no request routing, no usage dashboard and no billing. Anything you see labelled “Planned” on this site is a direction, not a shipped capability.