GPT-5.1

ModelActive

by OpenAI · family “gpt”

GPT-5.1 is the latest frontier-grade model in the GPT-5 series, offering stronger general-purpose reasoning, improved instruction adherence, and a more natural conversational style compared to GPT-5. It uses adaptive reasoning... (description from the OpenRouter listing)

in: imagein: textin: fileout: textReasoningTool useStructured output

Epoch Capabilities Index

149.7 #51 of 268

90% CI 148.2 – 151.0

Listing ↗

Released

13 Nov 2025

Context window

400K tokens

Max output

128K tokens

Input $ / 1M tokens

$1.25

Output $ / 1M tokens

$10.0

Cached input $ / 1M

$0.13

Each benchmark shown separately with its own source

Benchmark results (17)

All benchmarks

Benchmark	Domain	Score	vs best recorded	Setting	Run	Source
LMCA	agents	43.9%	67%	high	—	External ↗
DTBench	reasoning	90.1%	92%	high	—	External ↗
CL-bench Life	long-context	17.3%	78%	high	—	External ↗
CL-bench	long-context	23.7%	85%	high	—	External ↗
Chess Puzzles	games	32.0% ±4.7	44%	high	8 Dec 2025	Eval log ↗
SimpleQA Verified	knowledge	48.0% ±1.6	63%	high	27 Aug 2026	Eval log ↗
FrontierMath-Tier-4-2025-07-01-Privatesuperseded	math	12.5% ±4.8	26%	high	13 Nov 2025	Epoch ↗
GSO-Bench	coding	13.7%	29%	high	—	External ↗
ARC-AGI-2	reasoning	17.6%	19%	high	—	External ↗
FrontierMath-2025-02-28-Privatesuperseded	math	31.0% ±2.7	59%	high	13 Nov 2025	Epoch ↗
VPCT	multimodal	58.7%	65%	high	—	External ↗
WeirdML	coding	60.8% ±0.0	65%	high	—	External ↗
OTIS Mock AIME 2024-2025	math	88.6% ±4.0	89%	high	13 Nov 2025	Eval log ↗
SimpleBench	reasoning	53.2%	65%	high	—	External ↗
SWE-Bench verified	coding	68.0% ±2.1	81%	high	18 Feb 2026	Eval log ↗
GPQA diamond	science	87.6% ±1.9	91%	high	13 Nov 2025	Eval log ↗
ARC-AGI	reasoning	72.8%	74%	high	—	External ↗

Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.

API list price over time

Price history

$ per 1M tokens

11 observations (OpenRouter listing + Internet Archive snapshots). History accumulates with every ingest run; a single point means no change has been observed yet.