APIPureGamma API · OpenAI Compatible

Connect to the PureGamma API in three steps

Use your sk-pg key and this Base URL in any OpenAI-compatible client. PureGamma handles routing only to enabled providers; you do not manage upstream provider connections.

Open API Console Full API docs

Before you start

Create a key

Create an sk-pg-... key in the API Console. The plaintext key is shown once.

Add API balance

Gateway uses its own prepaid USD balance, separate from product Credits.

Copy configuration

Put the Base URL, key, and a returned model ID into your server or compatible client.

Gateway configuration card

Works with the OpenAI SDK, Dify, LangChain, Cursor, Continue, Open WebUI, and your own backend.

API type: OpenAI Compatible / OpenAI API
Base URL: https://api.puregamma.ai/v1
API key: sk-pg-... (environment variable or secret manager only)
Model: An exact id returned by GET /v1/models

.env

PUREGAMMA_API_KEY=sk-pg-...
PUREGAMMA_BASE_URL=https://api.puregamma.ai/v1
# Use an exact id returned by GET /v1/models
PUREGAMMA_MODEL=deepseek-v4-flash

Choose a model

These are public compatibility IDs. Your GET /v1/models response is the source of truth for availability.

deepseek-v4-flashDefaultLow-latency chat, batch work, and value-oriented tasks

deepseek-v4-proComplexMore demanding reasoning, code, and longer answers

kimi-k3-maxLong contextLong documents and complex tool workflows

glm-5.2GeneralGeneral Chinese/English tasks and tool calls

Model availability and JSON, streaming, or tool support may change with provider health and price approval. Request /v1/models first, then use an id it returns.

Run one request first

No SDK needed. Replace the environment variable with your own key and run this command.

curl

curl -sS https://api.puregamma.ai/v1/chat/completions \
  -H "Authorization: Bearer $PUREGAMMA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-flash",
    "messages": [{"role": "user", "content": "Hello from PureGamma."}]
  }'

Use the official OpenAI SDK

Minimal deployment · Keep the key server-side. Add these variables in Vercel, Railway, Docker, Cloud Run, or your own backend.

Python

from openai import OpenAI
import os

client = OpenAI(
    api_key=os.environ["PUREGAMMA_API_KEY"],
    base_url="https://api.puregamma.ai/v1",
)

response = client.chat.completions.create(
    model="deepseek-v4-flash",
    messages=[{"role": "user", "content": "Hello from PureGamma."}],
)
print(response.choices[0].message.content)

Node.js

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.PUREGAMMA_API_KEY,
  baseURL: "https://api.puregamma.ai/v1",
});

const response = await client.chat.completions.create({
  model: "deepseek-v4-flash",
  messages: [{ role: "user", content: "Hello from PureGamma." }],
});
console.log(response.choices[0].message.content);

Streaming and tools

Chat Completions supports stream: true, JSON mode, and tools. Check the selected model's capabilities before enabling them. Tool arguments are untrusted input: validate schema and permissions before they touch databases, payments, networks, or files.

Security and billing

Never place an sk-pg key in browser code, a mobile app, a repository, a screenshot, or a support ticket. Pause or rotate a suspected key in the API Console. Each request debits Gateway prepaid USD balance only; insufficient balance returns 402.

Need streaming, tool calls, error codes, or provider configuration for administrators? Open full documentation