How do I use Baseten models through the OpenRouter API?

OpenRouter serves 3 Baseten models behind one OpenAI-compatible API. Create an OpenRouter API key, point your client at https://openrouter.ai/api/v1, and set the model to an ID such as baseten/joey-1. The quickstart has request examples for every supported SDK.

Which Baseten model has the largest context window?

Joey 1 has the largest context window of any Baseten model on OpenRouter, accepting up to 16,384 tokens per request.

What is the newest Baseten model on OpenRouter?

Joey 1 is the most recently added Baseten model on OpenRouter, listed on October 29, 2026.

Baseten

Access 3 Baseten models through the OpenRouter unified API including Joey 1, Mammoth 1, and Elephant 1. Generate structured decisions and compare pricing, output modalities, and capabilities between different Baseten models.

Baseten tokens processed on OpenRouter

Baseten: Joey 1Joey 1
9.66B tokens
Joey 1 is a 2B decision model from Base Labs, Baseten's research team, built on Qwen3.5-2B. Given a state (text or JSON) and typed questions (yes/no, multiple choice, or ordinal score), it returns a calibrated probability for every option in a single forward pass, with no text generation. It is TypeSafe System One compatible and suits high-volume routing, classification, and guardrail checks where latency and cost matter most. Its larger siblings are Elephant 1 and Mammoth 1.
by basetenOct 29, 2026$0.01/M input tokens$0/M output tokens
Baseten: Mammoth 1Mammoth 1
8.41B tokens
Mammoth 1 is a 27B decision model from Base Labs, Baseten's research team, built on Qwen3.5-27B. Given a state (text or JSON) and typed questions (yes/no, multiple choice, or ordinal score), it returns a calibrated probability for every option in a single forward pass, with no text generation. It is TypeSafe System One compatible and suits classification, routing, and scoring where you need a scored answer instead of prose. Its smaller sibling is Elephant 1.
by basetenOct 21, 2026$0.10/M input tokens$0/M output tokens
Baseten: Elephant 1Elephant 1
6.86B tokens
Elephant 1 is a 9B decision model from Base Labs, Baseten's research team, built on Qwen3.5-9B. Given a state (text or JSON) and typed questions (yes/no, multiple choice, or ordinal score), it returns a calibrated probability for every option in a single forward pass, with no text generation. It is TypeSafe System One compatible and suits routing, classification, and scoring where you need a scored answer instead of prose. Its larger sibling is Mammoth 1.
by basetenOct 21, 2026$0.035/M input tokens$0/M output tokens

Baseten

Documentation

Frequently asked questions