Open GPT reasoning model for self-hosted agents and controllable deployments
https://openrouter.ai/api/v1 openai/gpt-oss-20b OpenAI: gpt-oss-20b (free) is listed for chat, coding workloads and supports a 131K context window.
Use the comparison and availability sections before choosing this listing, because it is not currently marked as free.
We only found this gpt-oss listing on OpenRouter in the current catalog. Use the related models section for nearby alternatives.
| Provider | Model listing | Access | Context | API | Limits |
|---|---|---|---|---|---|
| | OpenAI: gpt-oss-20b (free) | Paid | 131K | OpenAI-style | 200 req/day (free tier) |
| Model | Provider | Context | Access |
|---|---|---|---|
| Poolside: Laguna S 2.1 (free) | OpenRouter | 262K | Free tier |
| Poolside: Laguna XS 2.1 (free) | OpenRouter | 262K | Free tier |
| Cohere: North Mini Code (free) | OpenRouter | 256K | Free tier |
| NVIDIA: Nemotron 3.5 Content Safety (free) | OpenRouter | 128K | Free tier |
| NVIDIA: Nemotron 3 Ultra (free) | OpenRouter | 1.0M | Free tier |
OpenAI: gpt-oss-20b (free) is tagged for chat in this catalog and works with OpenAI-compatible client libraries.
OpenAI: gpt-oss-20b (free) is tagged for coding in this catalog and works with OpenAI-compatible client libraries.
OpenAI: gpt-oss-20b (free) is not currently marked as free on OpenRouter.
The model ID shown in this catalog is openai/gpt-oss-20b.
The listed free tier limit is 200 req/day (free tier). Limits can change per account tier, so confirm against the provider dashboard.
The listed context window is 131K tokens with up to 33K output tokens.
GPT-OSS 20B is OpenAI's open-weight 21B-parameter Mixture-of-Experts model with 3.6B active per forward pass, available free on OpenRouter (also on Groq, Cerebras, and Cloudflare). Supports reasoning level configuration, fine-tuning, and agentic capabilities — function calling, tool use, and structured outputs. Trained in OpenAI's Harmony response format. Released under Apache 2.0. The low active parameter count enables lower-latency inference on consumer or single-GPU hardware. 131K context window. Text-only. OpenAI-compatible. Free tier: 200 RPD on OpenRouter.
For API keys, setup steps, and provider-level limits, see the OpenRouter provider page.