> gpt-oss-20b is an open-weights AI model from OpenAI, released 5 Aug 2025. Index score 128 (GPT-4 = 100), rank 123 of 229 ranked models. API price $0.018 in / $0.090 out per million tokens, 131K context.

[Models](https://themodelindex.org/models/) /[OpenAI](https://themodelindex.org/labs/openai/)

# gpt-oss-20b

OpenAI

Released 5 Aug 2025

Open

Date verified

[Compare](https://themodelindex.org/compare/gpt-oss-20b-vs-gpt-5/) [Run it yourself →](https://themodelindex.org/local/) [Announcement](https://openai.com/index/gpt-oss-model-card/)

128116–139

Index score

#123 of 229 models

2/ 8

Core tests taken

2 of 6 areas · plus 5 anchor tests

−19

Versus the frontier at release

best then: o3

$0.018/ $0.090

Price per million tokens

input / output, list price

131K

Context window

tokens it reads at once

## Where it sits · on the frontier

_Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_

o1 reached this score 8 months earlier.

## Test results · by area · tick = best by any model

Show 5 anchor tests

Reasoning & knowledge

evidence

GPQA Diamond

61%

Humanity's Last Exam

no result

Maths

evidence

FrontierMath T1-3

no result

Coding

evidence

SWE-bench Verified

no result

Terminal-Bench

3%

Agents

evidence

APEX-Agents

no result

Novel problems

no evidence yet

ARC-AGI-2

no result

Long tasks

no evidence yet

METR time horizon

no result

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

## What others say · rank on each leaderboard

The Model Index

independent tests, our method

#123

of 219 · 128

Epoch Capabilities Index

overall capability from many benchmarks

#112

of 239 · 137.8

LMArena · Maths

maths questions in chat

#199

of 373 · 1335

Artificial Analysis Intelligence Index

overall capability across ten evaluations

#113

of 207 · 10

LMArena · Coding

coding questions in chat

#222

of 383 · 1369

LMArena · Text

overall chat quality, judged by people

#240

of 388 · 1318

LMArena · Creative writing

creative writing

#277

of 386 · 1240

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.

## Best at · top three among current models

Not in the top three among current models on any category we track.

Where this model places in the top three of today's models, on category leaderboards and on our core tests.

## Where to use it · prices checked 1 Oct 2026

#### From the lab

ChatGPT

app · subscription

↗

OpenAI API

API

↗

#### API providers · 11 · per million tokens, in / out

D

Darkbloom

$0.018 / $0.090 · fp8

cheapest

A

AkashML

$0.020 / $0.10 · fp4

↗

C

CoreWeave

$0.030 / $0.13 · fp4

↗

D

DekaLLM

$0.029 / $0.14 · bf16

↗

D

DeepInfra

$0.030 / $0.14 · bf16

↗

P

Parasail

$0.030 / $0.15 · fp4

↗

N

Novita

$0.040 / $0.15 · fp4

↗

S

SiliconFlow

$0.040 / $0.18 · fp8

↗

#### Run it yourself

🤗

Official weights

openai/gpt-oss-20b

↗

Provider prices come from OpenRouter's live listing; going direct to a provider can differ. Subscriptions are the lab's own apps.

## Facts and sources

Released

5 Aug 2025

Announcement

[openai.com ↗](https://openai.com/index/gpt-oss-model-card/)

Release entry

gpt-oss-120b/20b

Scored as

gpt-oss-20b

Artificial Analysis index

9 · high

Output speed

171 tokens/s

Hugging Face

[openai/gpt-oss-20b ↗](https://huggingface.co/openai/gpt-oss-20b)

## OpenAI releases · around this one

codex-1

Closed

16 MAY 2025

gpt-oss-120b

Open

5 AUG 2025

index 131

gpt-oss-20b

Open

5 AUG 2025

index 128

GPT-5

Closed

7 AUG 2025

index 154

GPT-5-Codex

Closed

15 SEP 2025

---
Source: https://themodelindex.org/models/gpt-oss-20b/ · The Model Index · data as of 2026-10-01