OpenAI: gpt-oss-20b API Pricing
by openai · 131K context window · #19 cheapest paid
Explore OpenAI's gpt-oss-20b API pricing for large language model inference. Designed for machine learning engineers, this page details the cost-effective solution for high-context applications. Input tokens are priced at $0.03000 per 1 million tokens, while output tokens cost $0.14000 per 1 million tokens, resulting in a total cost of $0.1700 per 1 million tokens. Benefit from a substantial 131,072 token context window. For example, processing 100 million tokens per day would cost approximately $510 per month. Compare OpenAI's gpt-oss-20b API with other LLM providers and estimate your potential savings.
Monthly Cost Examples
Assuming 50% input / 50% output token split
| Usage | Monthly cost |
|---|---|
| 100K tokens/month | <$0.01 |
| 1M tokens/month | $0.09 |
| 10M tokens/month | $0.85 |
| 100M tokens/month | $8.50 |
Compare with other models
OpenAI: gpt-oss-20b vs OpenAI: GPT-4o →OpenAI: gpt-oss-20b vs OpenAI: GPT-4o-mini →OpenAI: gpt-oss-20b vs OpenAI: o1 →Automate your model selection
StormRouter sends each request to the cheapest model that can handle it.
Only use OpenAI: gpt-oss-20b when your quality requirements demand it.