GPT OSS 20B
Groq first-party · 131,072 ctx · from $0.029 input / $0.14 output per 1M tokens
- Input price ranks #47 of 447 priced models (cheaper = better).
- Cheaper alternative with ≥ same context: Llama 3.1 8B Instruct at $0.02 input.
- Input price across curated sources ranges $0.029–$0.075 per 1M tokens.
| Provider | Groq |
|---|---|
| Family | gpt-oss |
| Context window | 131,072 |
| Max output | 65,536 |
| Input price | $0.029 |
| Output price | $0.14 |
| Cache read | — |
| Cache write | — |
| Reasoning | Yes |
| Tool calling | Yes |
| Vision | No |
| Open weights | Yes |
| Knowledge cutoff | |
| Released | 2025-08-05 |
| Last updated | 2025-09-25 |
Available from 2 curated sources
| Source | Input | Output |
|---|---|---|
| OpenRouter | $0.029 | $0.14 |
| Groq first-party | $0.075 | $0.3 |
Identity: Groq; data via models.dev + OpenRouter public APIs. Always confirm on the provider's official pricing page.