Patchwire

OpenAI · Released Oct 6, 2025

GPT-5 Pro

GPT-5 Pro 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...

ReasoningVision
  • Top 25%Chess Puzzles
  • Top 40%SimpleBench
  • Top 44%Humanity's Last Exam
Compare in showdown → OpenRouter ↗ All OpenAI models
Released
Oct 6 2025
Context window
400K 128K out
Input / 1M tokens
$15.00
Output / 1M tokens
$120.00

Where to run it

Providers

Only OpenAI serves GPT-5 Pro on OpenRouter right now.

ProviderInputOutputContextMax outputvs. best
OpenAI $15.00 $120.00 400K128K Cheapest

Value

Price vs. performance

8 models score higher for the same price or lesson Humanity's Last Exam, e.g. GPT-6 Astra.
8.1%23.7%39.2%54.8% $0.10$1.00$10.00$100.00 Claude Haiku 5.5 (Anthropic) — 30.1% · $0.10/1M in Claude Sonnet 5.5 (Anthropic) — 46.1% · $2.00/1M in Gemini 3.8 Flash (Google) — 44.5% · $0.82/1M in Claude Fable 5.1 (Anthropic) — 46.5% · $10.00/1M in Gemini 3.1 Flash Lite (Google) — 8.6% · $0.25/1M in Claude Opus 4.7 (Anthropic) — 36.2% · $5.00/1M in Gemini 3.1 Flash Lite Preview (Google) — 8.6% · $0.25/1M in Gemini 3.1 Pro Preview (Google) — 46.4% · $2.00/1M in Claude Opus 4.6 (Anthropic) — 34.4% · $5.00/1M in Kimi K2.5 (Moonshot AI) — 24.4% · $0.50/1M in Claude Opus 4.5 (Anthropic) — 25.2% · $5.00/1M in Claude Sonnet 4.5 (Anthropic) — 13.7% · $3.00/1M in Claude Opus 4.1 (Anthropic) — 11.5% · $15.00/1M in GLM 4.5 (Z.ai) — 8.3% · $0.60/1M in GLM 4.5 Air (Z.ai) — 8.1% · $0.13/1M in GPT-6 Astra (OpenAI) — 54.8% · $10.00/1M in GPT-5.4 Pro (OpenAI) — 44.3% · $30.00/1M in GPT-5.4 (OpenAI) — 36.2% · $2.50/1M in GPT-5.2 (OpenAI) — 27.8% · $1.75/1M in GPT-5.1 (OpenAI) — 23.7% · $1.25/1M in GPT-5 (OpenAI) — 25.3% · $1.25/1M in GPT-5 Mini (OpenAI) — 19.4% · $0.25/1M in GPT-5 Pro (OpenAI) — 31.6% · $15.00/1M in GPT-5 Pro

GPT-5 Pro Other OpenAI Other labs Value frontier · Humanity's Last Exam vs. input price per 1M tokens (log scale) · Drag to pan, pinch or Ctrl + scroll to zoom, double-click to zoom in.

Benchmarks

7 benchmark scores

BenchmarkScoreRankLeader
Humanity's Last ExamReasoning & knowledge · Expert questions across 100+ subjects 31.6% #10of 23 GPT-6 Astra54.8%
ARC-AGI-2Reasoning & knowledge · Few-shot abstract reasoning on grid puzzles 18.3% #46of 63 GPT-6 Astra95.0%
ARC-AGI-1Reasoning & knowledge · Pattern induction from input–output examples 70.2% #46of 63 Gemini 4 Argon98.5%
SimpleBenchReasoning & knowledge · Common-sense trick questions about space, time and people 61.6% #17of 43 Claude Fable 581.9%
Chess PuzzlesReasoning & knowledge · Find the single best move in 100 new puzzles 48.0% #17of 70 GPT-6 Astra76.0%
FrontierMathMaths · Expert-written problems, undergrad to early research (tiers 1–3) 55.8% #46of 72 GPT-6.1 Sol93.7%
FrontierMath Tier 4Maths · Exceptionally hard research-level problems 19.5% #47of 62 GPT-6.1 Sol100.0%

Specs

Details

Model ID
openai/gpt-5-pro
Developer
OpenAI
Release date
Oct 6, 2025 ·
Context window
400,000 tokens
Max output
128,000 tokens
API price
$15.00 / $120.00 per 1M input / output tokens
Providers
1
Input types
image, text, file
Reasoning
Yes
Tool calling
Yes
Open weights
No
Free tier
No

Release date, prices and providers from OpenRouter's catalog; buying directly from a provider can cost a little differently. Uptime is the share of OpenRouter requests each provider served over the last day. Benchmark scores are the best result per benchmark from Epoch AI, licensed CC BY 4.0. Rank counts every model tracked here with a score.