Developer docs
One API for open models.
An OpenAI-compatible endpoint for DeepSeek, MiniMax H3 and BytePlus video, with one key and a per-key budget. Every request in these docs can be run from the page.
deepseek-flash
DeepSeek Flash · text + image
deepseek-v4-pro
DeepSeek Pro · reasoning
minimax-h3
MiniMax H3 · video
minimax-h3-max
MiniMax H3-Max · video
seedance
BytePlus Seedance 2.5 · video
A deployment only serves groups whose provider credential is configured. Check GET /v1/models for the groups your key can call.
OpenAI-compatible
Point any OpenAI SDK at /v1. Chat, streaming, vision and video tasks share one bearer token.
Async video
Create, poll and download H3 and Seedance jobs at /v1/videos from these docs.
Metered
Per-key budgets, model allowlists and spend tracked against each key. A key's grants are frozen when it is issued.
First request
curl -sS "https://llms.party/v1/models" \
-H "Authorization: Bearer $LLMS_PARTY_KEY"const response = await fetch('https://llms.party/v1/models', {
headers: { authorization: `Bearer ${process.env.LLMS_PARTY_KEY}` },
});
const { data } = await response.json();
console.log(data);import os
import requests
response = requests.get(
"https://llms.party/v1/models",
headers={"Authorization": f"Bearer {os.environ['LLMS_PARTY_KEY']}"},
timeout=30,
)
print(response.json()["data"])The response lists the model groups your key can call. The inline playgrounds run the same check.
A deployment serves only the groups whose provider credentials it has, so this response is the authoritative list for your key. The pages below describe the full set of groups.