How much does gpt-oss-20b (free) cost?

gpt-oss-20b (free) costs $0.00 per million input tokens and $0.00 per million output tokens. For a typical conversation (~2,000 tokens), that's approximately $0.000 per message.

What benchmarks has gpt-oss-20b (free) been tested on?

gpt-oss-20b (free) has been evaluated on 9 benchmarks. Top scores: OpenCompass — IFEval: 88.9, OpenCompass — AIME2025: 87.9, OpenCompass — MMLU-Pro: 72.8.

Is gpt-oss-20b (free) open source?

Yes, gpt-oss-20b (free) is open source.

How does gpt-oss-20b (free) compare to Kimi K2 Thinking?

gpt-oss-20b (free) has an average score of 61.0 while Kimi K2 Thinking scores 61.0. Kimi K2 Thinking slightly outperforms gpt-oss-20b (free) overall. gpt-oss-20b (free) costs $0.00/1M input vs Kimi K2 Thinking at $0.60/1M input. See full comparison →

Home/Models/gpt-oss-20b (free)

gpt-oss-20b (free)

Name: gpt-oss-20b (free)
Author: OpenAI

by OpenAI · Released Aug 2025

Open Source

61.0

avg score

Rank #64

Compare

Better than 73% of all models

Context

131K tokens (~66 books)

Input $/1M

Free

Output $/1M

Free

Type

text

License

Open Source

Benchmarks

9 tested

Data updated today

About

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...

Tested on 9 benchmarks with 66.4% average. Top scores: OpenCompass — IFEval (88.9%), OpenCompass — AIME2025 (87.9%), OpenCompass — MMLU-Pro (72.8%).

Capabilities

coding

68.4

#20 globally

math

87.9

#8 globally

knowledge

51.1

#94 globally

speed

39.2

#43 globally

language

88.9

#17 globally

Benchmark Scores

Compare All

Tested on 9 benchmarks · Ranked across 5 categories

Score Distribution (all 233 models)

0255075100

▲ You are here

codingCompare coding →

OpenCompass — LiveCodeBenchV6

OpenCompass Live Code Bench v6. Fresh competitive programming problems to evaluate code generation without memorization.

68.4—

mathCompare math →

OpenCompass — AIME2025

OpenCompass evaluation on AIME 2025 problems. Tests mathematical reasoning on fresh competition problems.

87.9—

knowledgeCompare knowledge →

OpenCompass — MMLU-Pro

OpenCompass MMLU-Pro evaluation. Harder knowledge test with more answer choices.

72.8—

OpenCompass — GPQA-Diamond

OpenCompass evaluation of GPQA Diamond. PhD-level science questions from the hardest subset.

68.9—

OpenCompass — HLE

OpenCompass evaluation of Humanitys Last Exam. Expert-level cross-discipline knowledge test.

11.6—

Quick compare:

vs Kimi K2 Thinking

vs o3 Pro

vs Qwen3 Max

Excellent (85+) Good (70-85) Average (50-70) Below (<50)

Similar Models

Links

Info

OpenAI Pricing explorer Developers · API

Research

Documentation

Community

Source Code

BenchGecko API

gpt-oss-20b-free

Specifications

Typetext
Context131K tokens (~66 books)
ReleasedAug 2025
LicenseOpen Source
StatusActive
Cost / Message~$0.000

Available On

OpenAIFree

Frequently Asked Questions

gpt-oss-20b (free) is an open-source text AI model by OpenAI, released in August 2025. It has an average benchmark score of 61.0. Context window: 131K tokens.

Benchmarks

OpenCompass — IFEval OpenCompass — AIME2025 OpenCompass — MMLU-Pro OpenCompass — GPQA-Diamond OpenCompass — LiveCodeBenchV6

OpenAI · Provider OpenAI · Economy All Models Compare Models Pricing Developers · API Context Window · Glossary

gpt-oss-20b (free)

Frequently Asked Questions

Related Models

Benchmarks

Related Pages