gpt-oss-20b
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...
gpt-oss-20b ranks #15 of 377 models in the default AI Benchmark Hub order (quality first, then cost). At $0.05 blended per million tokens, it is cheaper than 96% of models with published API pricing. Its 131,072-token context window is 2.0× smaller than the catalog median (262,144 tokens).
Context
Max output: 32768
Pricing
Output / 1M: 0.09
Blend / 1M: 0.05
Quality
Provider
Moderated: no
Monthly cost at this blended rate
Estimated API spend if every token is billed at $0.05per million (the blend of this model's input and output prices). Real bills depend on the input/output mix.
| Tokens / month | Est. cost (USD) |
|---|---|
| 1,000,000 | $0.05 |
| 10,000,000 | $0.54 |
| 100,000,000 | $5.40 |
Related models
Same lab, similar price, or similar quality — useful next comparisons.
More from OpenAI
- gpt-oss-120b$0.06 / 1M
- gpt-oss-safeguard-20b$0.19 / 1M
- GPT-5 Nano$0.22 / 1M
- GPT-4.1 Nano$0.25 / 1M
- GPT-6 Luna Pro$0.30 / 1M
Closest API price
- gpt-oss-120b$0.06 / 1M
- GLM Flash Latest$0.06 / 1M
- Nex-N2.5-Mini$0.06 / 1M
- Llama 3 8B Lunaris$0.04 / 1M
- Mistral Small 3$0.07 / 1M