RSS

OpenAI

gpt-oss-20b

Benchmarks, price, context and versions

Also searched as

gpt oss 20b

Direct answer: gpt-oss-20b is a model from OpenAI. 60.8% on GPQA Diamond, per Epoch AI. On OpenRouter, US$ 0.03 per 1 million input tokens and US$ 0.13 per 1 million output tokens.

✓ Good for

  • high-volume use (input price in the cheapest 25% on OpenRouter)
  • running on your own infrastructure (open weights)

✕ Not the best for

  • technical reasoning on science topics (#123 of 198 on GPQA Diamond)
  • calculation and problems with an exact answer (#111 of 186 on OTIS Mock AIME)
  • technical reasoning as the model appears through the API (#106 of 135 on GPQA Diamond (OpenRouter))
  • agents that follow rules and use tools (#106 of 131 on τ-bench Verified (airline))

This block follows a rule of this section, using source data: if the model is in the top 25% of the metric, the item goes under "Good for"; if it is in the bottom half, it goes under "Not the best for". The middle band appears in neither. Price and context use OpenRouter; scores use Epoch AI or OpenRouter Benchmarks, in the latest version we checked.

Benchmarks

GPQA Diamond

?what it is
60.8%

highest score on this metric 96.4%

123º / 198 · Epoch AI, latest version we checked

OTIS Mock AIME

?
65.3%

highest score on this metric 100%

111º / 186 · Epoch AI, latest version we checked

GPQA Diamond (OpenRouter)

?
68.7%

highest score on this metric 96.5%

106º / 135 · OpenRouter Benchmarks, latest version we checked

τ-bench Verified (airline)

?
53.5%

highest score on this metric 82%

106º / 131 · OpenRouter Benchmarks, latest version we checked

Key specs

Origin
the US
Family
gpt oss provisional data
Weights?
open
License
no data
Input US$/1M?
US$ 0.03
Output US$/1M
US$ 0.13
Context?
131,072 tokens
Parameters
no data
Input
text
OpenRouter usage?
no data
Downloads (30 days)?
no data
Use cases provisional data
Text Agents
Dates in sources
  • 2025, published, per Epoch AI
  • 2025, listed on OpenRouter

Version history

Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch.

FAQ

What is gpt-oss-20b?

gpt-oss-20b is a model from OpenAI, with open weights, used for text, agents.

How much does gpt-oss-20b cost?

On OpenRouter, US$ 0.03 per 1 million input tokens and US$ 0.13 per 1 million output tokens.

What is the context window of gpt-oss-20b?

131,072 tokens, per OpenRouter.

Is gpt-oss-20b open source?

Yes, it has open weights. License: no data.

What is gpt-oss-20b's GPQA Diamond score?

60.8% on GPQA Diamond, per Epoch AI.

Is gpt-oss-20b good for coding?

No data: there is no SWE-bench Verified measurement for it in open sources.

Who made gpt-oss-20b?

It was made by OpenAI, a lab from the US.

provisional data Questions under review: Semrush had no API balance during collection. Answers use source data only.

Models from OpenAI

See lab

Sources

  • Epoch AI, CC BY 4.0, latest version we checked.
  • OpenRouter, dados públicos da API; exports de benchmarks e rankings em CC BY 4.0, latest version we checked.
  • OpenRouter Benchmarks, CC BY 4.0, latest version we checked.

Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch. Family and version were grouped by rule from the name.