gpt-oss-20b
OpenAIgpt-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...
In
$0.03
Out
$0.13
Blended
$0.06
Where to buy this model
10 access routes| Route | In | Out | Blended | vs |
|---|---|---|---|---|
| ● PrimeIntellect cheapestfp4 | $0.01 | $0.03 | $0.01 | - |
| ● Darkbloom fp8 | $0.02 | $0.09 | $0.04 | +300% |
| ● AkashML fp4 | $0.02 | $0.10 | $0.04 | +300% |
| ● CoreWeave fp4 | $0.03 | $0.13 | $0.06 | +500% |
| ● DekaLLM bf16 | $0.03 | $0.14 | $0.06 | +500% |
| $0.03 | $0.14 | $0.06 | +500% | |
| $0.03 | $0.15 | $0.06 | +500% | |
| ● SiliconFlow fp8 | $0.04 | $0.18 | $0.08 | +700% |
 Amazon Bedrock | $0.07 | $0.15 | $0.09 | +800% |
| $0.07 | $0.30 | $0.13 | +1200% |
● official API · ● cloud platform · ● inference host · USD per 1M tokens. Same model, different providers - price and quantization vary.
- Ctx
- 131K
- max output
- 118K
- cache read /1M
- $0.03
- Family
- gpt
- Modalities
- text
- Features
- json · reasoning · tools
Live pricing, updated 2026-11-03.
OpenAI models at a similar price
| Model | In | Out | Ctx |
|---|---|---|---|
| gpt-oss-120b | $0.024 | $0.087 | 131K |
| gpt-oss-20b · this model | $0.03 | $0.13 | 131K |
| gpt-oss-safeguard-20b | $0.075 | $0.3 | 131K |
| GPT-5 Nano | $0.05 | $0.4 | 400K |
| GPT-4.1 Nano | $0.1 | $0.4 | 1.047576M |
| GPT-6 Luna | $0.1 | $0.5 | 1.05M |