OpenAI

gpt-oss-20b

openai/gpt-oss-20b

JSON

Specification

Published by the vendor; the platform bills at the target's declared price.

gpt-oss-20b

reasoning open weights tools

OpenAI openai/gpt-oss-20b

gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for...

Input $/1M: $0.03
Output $/1M: $0.13
Context: 131K
Max output: 118K
Released: Aug 5, 2025
Modalities: text

Intelligence index-

Coding index21

Agentic index1

Cached input reads $0.03 per 1M · weights on Hugging Face as openai/gpt-oss-20b

On this deployment

No routing target resolves to this model yet. Declare one and the router picks it up without retraining.

Not routable here yet

Add a target with metadata.model set to this id.

targets.yamlyaml

# targets.yaml - declare a target that runs gpt-oss-20b
targets:
  - id: gpt-oss-20b
    kind: llm
    name: gpt-oss-20b
    metadata:
      model: openai/gpt-oss-20b          # links the target to this catalogue entry
    capabilities:
      domains: [general]
      supports_tools: true
      context_window: 131072
    cost:
      usd_per_1k_tokens: 0.000080

Price among OpenAI models

Input list price per million tokens; gpt-oss-20b is the cheapest of 13 listed.

More from OpenAI

Other models by the same vendor in the catalogue.