OpenAI
Frontier

GPT-5

Frontier

OpenAI's 2025 flagship, unified reasoning + multimodal model with a 400K context. Replaces the GPT-4 line as the default ChatGPT and API model.

GPT-5 is a frontier AI model from OpenAI. It costs $1.25 per million input tokens and $10.00 per million output tokens (blended $3.88/M), with a 400,000-token context window.

RatesUSD per million tokens. Blend is 70% input, 30% output, our calculation.

INPUT
$1.25/M
per million input tokens
OUTPUT
$10.00/M
per million output tokens
BLENDED 70/30
$3.88/M
unchanged since 3 May
CONTEXT
400,000
tokens

Hand-maintained from OpenAI's published pricing; no check date on record.

What it is good at
  • Unified reasoning + chat
  • 400K context
  • Native vision
  • Strong tool use
  • Best-in-class JSON mode
Typical use cases
  • General chat assistants
  • Vision tasks
  • Complex agentic workflows
  • Structured output generation

Benchmarks

vs. best public score
MMLU91%
Multitask academic knowledge across 57 subjects. No source recorded yet.
Graduate-level science questions, "Google-proof". Source: OpenAI (self-reported), GPQA diamond (no tools), read 2026-11-10.
Real GitHub issues solved end-to-end. Source: OpenAI (self-reported), SWE-bench Verified, read 2026-11-10.
LMArena Elo1434 Elo
Crowd-sourced head-to-head preference Elo rating. Source: LMArena (independent), LMArena Text Arena score, read 2026-11-10.
Hand-curated from each provider's published reports and public leaderboards. Methodology varies across sources, treat as directional rather than authoritative.

More from OpenAI

See all 28 →

Frequently asked questions

How much does GPT-5 cost?

GPT-5 costs $1.25 per million input tokens and $10.00 per million output tokens, for a blended reference rate of $3.88 per million tokens.

What is GPT-5's context window?

GPT-5 supports up to 400,000 tokens of context in a single request.

What is GPT-5 best for?

GPT-5 is well suited to Unified reasoning + chat, 400K context and Native vision.

Who makes GPT-5?

GPT-5 is developed and served by OpenAI. It was released in 2025.

Compared with

Terms used on this page