# GPT OSS 20B

> GPT OSS 20B is an open-weight language model from OpenAI, released on 5 Aug 2025. It accepts text and generates text, with a 131K-token context window and up to 16K output tokens per response. As of 4 Oct 2026, the lowest listed API price is $0.02 per million input tokens and $0.09 per million output tokens (OpenRouter), across 10 providers we track.

Source page: https://www.aimodel.directory/models/gpt-oss-20b
Last verified: 4 Oct 2026

## Specifications

- **Developer:** OpenAI
- **Release date:** 5 Aug 2025 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/openai/gpt-oss-20b.toml)
- **Status:** Generally available (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/openai/gpt-oss-20b.toml)
- **Weights:** Open weights (source: Hugging Face, https://huggingface.co/openai/gpt-oss-20b)
- **License:** apache-2.0 (source: Hugging Face, https://huggingface.co/openai/gpt-oss-20b)
- **Parameters:** 20.9B (source: Hugging Face, https://huggingface.co/openai/gpt-oss-20b)
- **Context window:** 131K (131,072 tokens) (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/openai/gpt-oss-20b.toml)
- **Max output:** 16K tokens (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/openai/gpt-oss-20b.toml)
- **Knowledge cutoff:** Jun 2024 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/openai/gpt-oss-20b.toml)
- **Input:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/openai/gpt-oss-20b.toml)
- **Output:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/openai/gpt-oss-20b.toml)
- **Reasoning mode:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/openai/gpt-oss-20b.toml)
- **Tool calling:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/openai/gpt-oss-20b.toml)
- **Structured output:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/openai/gpt-oss-20b.toml)
- **Hugging Face:** openai/gpt-oss-20b (source: Hugging Face, https://huggingface.co/openai/gpt-oss-20b)

## API pricing (USD per 1M tokens)

| Provider | Input | Output | Cached input | Context |
|---|---|---|---|---|
| NVIDIA NIM | Free | Free | – | 131K |
| OpenRouter | $0.02 | $0.09 | $0.009 | 131K |
| DeepInfra | $0.03 | $0.14 | – | 131K |
| Vercel AI Gateway | $0.03 | $0.14 | – | 131K |
| Novita AI | $0.04 | $0.15 | – | 131K |
| SiliconFlow | $0.04 | $0.18 | – | 131K |
| Amazon Bedrock | $0.07 | $0.30 | – | 131K |
| Amazon Bedrock | $0.07 | $0.30 | – | 131K |
| Groq | $0.07 | $0.30 | $0.04 | 131K |
| Amazon Bedrock | $0.08 | $0.36 | – | 128K |
| Hugging Face Inference Providers | $0.10 | $0.50 | – | 131K |
| Cloudflare Workers AI | $0.20 | $0.30 | – | 128K |

## FAQ

### What is the context window of GPT OSS 20B?

GPT OSS 20B has a context window of 131,072 tokens (131K) and can generate up to 16,384 tokens in a single response, according to models.dev.

### How much does the GPT OSS 20B API cost?

GPT OSS 20B is listed from $0.02 per million input tokens and $0.09 per million output tokens (OpenRouter). Prices last verified 4 Oct 2026.

### Is GPT OSS 20B open source?

GPT OSS 20B is an open-weight model: its weights are published on Hugging Face as openai/gpt-oss-20b under the apache-2.0 license. Check the license terms before commercial use.

### When was GPT OSS 20B released?

OpenAI released GPT OSS 20B on 5 Aug 2025, according to models.dev.

### Which providers offer GPT OSS 20B?

We track GPT OSS 20B on 10 providers: NVIDIA NIM, OpenRouter, DeepInfra, Vercel AI Gateway, Novita AI, SiliconFlow, Amazon Bedrock, Groq, Hugging Face Inference Providers, Cloudflare Workers AI.

---
Data licensed CC BY 4.0. Cite as: AI Model Directory, "GPT OSS 20B", https://www.aimodel.directory/models/gpt-oss-20b