Model

GPT-5.1

CLOSED

OpenAI · GPT-5.1 family · released Nov 12, 2025

Warmer, more conversational refresh with adaptive reasoning; Instant and Thinking modes.

ReasoningCodingVisionFunction callingTool useAgentic

2215.2

Elo · rank #60

Parameters

Undisclosed

Active params

—

Context

400K tokens

Architecture

GPT-5 upgrade with adaptive reasoning spanning Instant and Thinking behavior in one model; dense transformer, exact size undisclosed

License

Proprietary (OpenAI API Terms)

Languages

—

API price (in/out)

$1.25 / $10

Modalities

text · vision

Benchmark results

Bar shows position within the tracked field; marker = field best

AIMEMath94.0%#24

best: GPT-5.2 · 100.0%

ARC-AGI-1Reasoning72.8%#11

best: GPT-6 Astra · 98.5%

ARC-AGI-2Reasoning17.6%#23

best: GPT-6 Astra · 95.0%

GPQA DiamondReasoning88.1%#43

best: GPT-6 Astra · 96.0%

Humanity's Last ExamReasoning23.7%#66

best: Claude Opus 5 · 64.7%

LiveCodeBenchCoding86.5%#20

best: DeepSeek-V4-Pro (Think Max) · 93.5%

MMMUVision85.4%#7

best: Claude Fable 5 · 89.3%

SWE-bench VerifiedCoding76.3%#40

best: Claude Opus 5 · 96.0%

Run it locally

Closed weights — available via API only. No local deployment.

Input / M tok

$1.25

Output / M tok

$10

GPT-5.1 family

Elo progression across releases

GPT-5.1Nov 20252215.2

Resources

DOCSdevelopers.openai.com/api/docs/models/gpt-5.1 APIAvailable via OpenAI API

API price $1.25/$10 · each benchmark row carries its own source badge (see methodology)