GPT-5

Name: GPT-5 model facts
Creator: OSMEXO

openai/gpt-5

OpenAI
Family · GPT-5
Released 7 Aug 2025
Text · Image · File → Text
Reasoning · adjustable
Tools · JSON · Vision

Not on OSMEXO yet

Request this model

Spec & list price

Context391K

Max output128K

Input / 1M$1.25

Output / 1M$10.008× input

Scores · standing

Intelligence23.0#110/489

Coding37.8#95/196

Usage

Tokens / day716.6M▼ 20% 7d

Usage rank#205

Requests / day82.2K

Tool calls / req0.09

Per request · 30 days

Avg input5.5K0.4× median

Avg output9221.3× median

Cache hit56%0.8× median

Reasoning share74%1.3× median

Tool-call failures0.6%0.3× median

Consumption

Peak 2.7B 22 OctLow 508.0M 25 OctRank today #205

News & trends

44 mentions this week ▲ 175%this model · videos, posts, articles · 30 days

Trending · 14 days

News · latest from OpenAI

GPT-6.1 Terrasimonwillison.net · 31 Oct
GPT-5.6 Sol can beat Factorio on peaceful moderyanmadden.net · 31 Oct
Post-training 4B model with RL to fly Tesla Roadster 21% better than GPT-6 Astraskypilot.ai · 30 Oct
DeepSeek, Moonshot used OpenAI's GPT-6 Astra to train latest modelsreuters.com · 29 Oct

Benchmarks

bar = share of the top score · Benchmark data: Artificial Analysis

AIME#1 of 160

95.7%

MATH-500#1 of 167

99.4%

MMLU-Pro#9 of 273

87.1%

AIME 2025#11 of 206

94.3%

Math index#11 of 206

94.3

LiveCodeBench#14 of 269

84.6%

GPQA Diamond#91 of 445

85.4%

Long-context reasoning#76 of 393

78.2%

Speed · median across 3 providers, latest day

Output speed#177/33362 tok/sabout median

Fastest provider78 tok/sbest of 3 providers

First token#3053.68 s3.6× longer than median

#29 cheapest of 33 with a similar score

Value

Intelligence Index

Blended price per 1M tokens (log)

Family progress

Intelligence Index

Family releases38.4

Technical details

Reasoning: on · medium
Tokenizer: GPT
Knowledge cutoff: 2024-09-30
Moderation: moderated

Reasoning effort

high
medium · default
low
minimal

Request parameters 11

tools
tool_choice
response_format
structured_outputs
reasoning
reasoning_effort
include_reasoning
max_tokens
max_completion_tokens
seed

+1 more

verbosity

Interesting facts

Less output room than similar models (128.0K against a 235.9K median).
Weekdays average 1.7B tokens a day against 753.7M on weekends.

When to use

Best for

General-purpose assistants with image and file input
Coding and math-heavy reasoning
Long document Q&A
Tool-calling agents with structured output

Think twice for

Image or audio generation
Self-hosting (closed weights)
High-volume simple chat (cheaper options exist)

Research papers

Same family