Polarison

OpenAI · open weights

gpt-oss-20b API pricing by provider

gpt-oss-20b is available from 13 providers. The cheapest is Darkbloom at $0.02 input / $0.10 output per 1M tokens (FP8). The priciest host, Groq, charges 3.3x the cheapest.

Prices from the OpenRouter API, retrieved September 15, 2026. Weights: openai/gpt-oss-20b.

ProviderPrecisionInput / 1MOutput / 1MDocument Q&A / monthContextMax output
Darkbloomfp8$0.02$0.10$11.00131K33K
AkashMLfp4$0.02$0.10$11.00131K118K
CoreWeavefp4$0.03$0.13$15.90131K118K
DekaLLMbf16$0.029$0.14$15.80131K118K
DeepInfrabf16$0.03$0.14$16.20131K118K
Parasailfp4$0.03$0.15$16.50131K118K
Phala—$0.04$0.15$20.50131K118K
Novitafp4$0.04$0.15$20.50131K33K
SiliconFlowfp8$0.04$0.18$21.40131K8K
Together—$0.05$0.20$26.00131K118K
Amazon Bedrock—$0.07$0.15$32.50131K118K
Google—$0.07$0.25$35.50131K33K
Groq—$0.075$0.30$39.00131K66K

Sorted by blended price (3 input : 1 output). Document Q&A assumes 50,000 questions a month with 8,000 input and 600 output tokens each. Precision is reported by the provider; “—” means not disclosed.

Frequently asked questions

What is the cheapest gpt-oss-20b API provider?
Darkbloom is the cheapest healthy provider we track at $0.02 per 1M input tokens and $0.10 per 1M output tokens.
How much does gpt-oss-20b cost?
Across 13 providers, input prices range from $0.02 to $0.075 per 1M tokens and output prices from $0.10 to $0.30.
What is gpt-oss-20b’s context window?
gpt-oss-20b supports up to 131K tokens, but some providers serve a smaller context window — check the table before choosing.

Other open-weight models