GPT-5

ModelActive

by OpenAI · family “gpt”

GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and accuracy... (description from the OpenRouter listing)

in: textin: imagein: fileout: textReasoningTool useStructured output

Epoch Capabilities Index

150.0 #49 of 268

90% CI 0.0 – 0.0

Listing ↗

Released

7 Aug 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 (25)

All benchmarks

Benchmark	Domain	Score	vs best recorded	Setting	Run	Source
LMCA	agents	40.0%	61%	high	—	External ↗
DTBench	reasoning	90.7%	92%	high	—	External ↗
Mystery Game Puzzles	games	23.0% ±4.2	27%	high	5 Aug 2026	Epoch ↗
EBR-bench	reasoning	12.7%	17%	high	29 Jun 2026	Epoch ↗
FrontierMath-Tiers-1-3-v2-Private	math	55.4% ±2.9	59%	high	10 Jun 2026	Epoch ↗
FrontierMath-Tier-4-v2-Private	math	22.0% ±6.5	22%	high	11 Jun 2026	Epoch ↗
ProofBench	math	18.0%	18%	high	—	External ↗
Chess Puzzles	games	37.0% ±4.9	51%	high	8 Dec 2025	Eval log ↗
SimpleQA Verified	knowledge	50.1% ±1.6	66%	high	27 Aug 2026	Eval log ↗
FrontierMath-Tier-4-2025-07-01-Privatesuperseded	math	12.5% ±4.8	26%	high	30 Oct 2025	Epoch ↗
DeepResearch Bench	agents	48.1%	87%	high	—	External ↗
GSO-Bench	coding	6.9%	15%	high	—	External ↗
ARC-AGI-2	reasoning	9.9%	10%	high	—	External ↗
METR Time Horizons	agents	69.4%	81%	high	—	External ↗
FrontierMath-2025-02-28-Privatesuperseded	math	32.4% ±2.8	62%	high	13 Nov 2025	Epoch ↗
VPCT	multimodal	66.0%	73%	high	—	External ↗
HLE	knowledge	25.3%	46%	high	—	External ↗
WeirdML	coding	60.7% ±0.0	65%	high	—	External ↗
Aider polyglot	coding	88.0%	100%	high	—	External ↗
OTIS Mock AIME 2024-2025	math	91.4% ±3.7	91%	high	29 Oct 2025	Eval log ↗
SimpleBench	reasoning	56.7%	69%	high	—	External ↗
SWE-Bench verified	coding	73.6% ±2.0	88%	high	6 Feb 2026	Eval log ↗
GPQA diamond	science	86.2% ±2.1	90%	high	29 Oct 2025	Eval log ↗
MATH level 5	math	98.1% ±0.3	100%	high	29 Oct 2025	Epoch ↗
ARC-AGI	reasoning	65.7%	67%	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

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