100% open-weight models

The unified interface for open LLMs

Better prices, better uptime, no lock-in. One OpenAI- and Anthropic-compatible API for Kimi, MiniMax, Qwen, DeepSeek, GLM and GPT-OSS — routed across live inference providers at the edge.

Browse models Quickstart
40+
Curated open models
13
Live providers
203
Live routes
1M
Max context window
curl · /oss/v1/chat/completions 
# OpenAI-compatible — just swap the base URL
curl https://api.opengateway.one/oss/v1/chat/completions \
  -H "Authorization: Bearer $OPENGATEWAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "moonshotai/Kimi-K2.7-Code:preferred",
    "messages": [{"role": "user", "content": "Refactor this function"}]
  }'
Why OpenGateway

One API. Every open model.

Access the strongest open-weight models through a single, unified interface. The OpenAI and Anthropic SDKs work out of the box.

One API for any open model

Call Kimi, Qwen, DeepSeek, GLM, MiniMax and GPT-OSS through one endpoint. Drop-in compatible with the OpenAI and Anthropic SDKs — change the base URL and go.

Higher availability

Every model is hosted by multiple inference providers. OpenGateway routes to a healthy one and fails over automatically when a provider degrades — across 13 live providers.

Price & performance

Pick a routing policy per request — :cheapest, :fastest or :preferred. Runs at the edge for minimal latency between your users and inference.

100% open weights

No frontier lock-in. Every model on OpenGateway is open-weight and self-hostable — so you keep portability, control over your data, and freedom to leave.

Featured models

Flagship open weights

View all models →
Explore

Browse by capability

Quickstart

Start in 30 seconds

The OpenAI SDK works by changing one line. Anthropic clients work too via the Messages API — point them at /oss.

python · openai sdk 
from openai import OpenAI

client = OpenAI(
    base_url="https://api.opengateway.one/oss/v1",
    api_key="$OPENGATEWAY_API_KEY",
)

resp = client.chat.completions.create(
    model="moonshotai/Kimi-K2.7-Code:preferred",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)
typescript · openai sdk 
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.opengateway.one/oss/v1",
  apiKey: process.env.OPENGATEWAY_API_KEY,
});

const r = await client.chat.completions.create({
  model: "MiniMaxAI/MiniMax-M3:preferred",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(r.choices[0].message.content);
Route policies

Append a policy to any model id to control routing.

:preferred — best overall route :fastest — lowest latency :cheapest — lowest price
open weights · no lock-in

Ship on open models today

Browse the catalog, copy a model id, and send your first request. Every model is open-weight and routed across live providers.

Browse models View live status