Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d271Policy30d66, 14 pendingSources1,015

GPT-5 nano

OpenAIReleased Aug 7, 2025Current

Score basis

GPT-5 nano is behind the leaders in instruction following, multimodal tasks, agentic tasks, coding, and math. Too few results yet to rate reasoning, safety, long context, multilingual tasks, or factuality.

Capability profile

Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.

Limited5 capabilities
  1. Instruction Following−30.5%1 of 3
  2. Multimodal−36.5%1 of 6
  3. Agentic−41.3%1 of 7
  4. Coding−42.8%3 of 10
  5. Math−54.9%2 of 5
Not rated5 capabilities

Too few results yet to rate Reasoning, Safety, Long Context, Multilingual or Factuality.

Price

$0.05input$0.40outputper million tokens

From Artificial Analysis · 2 providers tracked · All prices

Cheaper than 86% of 330 priced models · 3:1 input-to-output blend, log scale

Evidence

85results on43benchmarks

  • 34 independently verified
  • 46 aggregator
  • 5 vendor-reported

From 21 sources · latest Oct 8, 2026 · How verification works

API features

Tool callingStructured outputsReasoning

As listed by OpenRouter

Research

12 papers reference GPT-5 nano

GPT-5 nano benchmark results

85 results on 43 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Instruction Following

LimitedFull instruction following ranking

30.5% behind the leader1 of 3 ranked benchmarks measured

Show 2 more instruction following resultsHide 2 instruction following results

Multimodal

LimitedFull multimodal ranking

36.5% behind the leader1 of 6 ranked benchmarks measured

Show 4 more multimodal resultsHide 4 multimodal results

Agentic

LimitedFull agentic ranking

41.3% behind the leader1 of 7 ranked benchmarks measured

Show 2 more agentic resultsHide 2 agentic results

Coding

LimitedFull coding ranking

42.8% behind the leader3 of 10 ranked benchmarks measured

Show 4 more coding resultsHide 4 coding results

Math

LimitedFull math ranking

54.9% behind the leader2 of 5 ranked benchmarks measured

Show 2 more math resultsHide 2 math results

Reasoning

Not enough dataFull reasoning ranking

0 of 6 ranked benchmarks measured

Show 9 more reasoning resultsHide 9 reasoning results

Long Context

Not enough dataFull long context ranking

0 of 3 ranked benchmarks measured

Factuality

Not enough dataFull factuality ranking

0 of 4 ranked benchmarks measured

Show 5 more factuality resultsHide 5 factuality results

More results

Benchmarks outside the capability baskets. They are not ranked against a leader.

Show 34 more resultsHide 34 results

GPT-5 nano: common questions

Who makes GPT-5 nano?

GPT-5 nano is made by OpenAI.

When was GPT-5 nano released?

GPT-5 nano was released on Aug 7, 2025, according to Artificial Analysis.

What is GPT-5 nano good at?

GPT-5 nano is behind the leaders in instruction following, multimodal tasks, agentic tasks, coding, and math. Too few results yet to rate reasoning, safety, long context, multilingual tasks, or factuality.

How much does GPT-5 nano cost?

GPT-5 nano costs $0.05 per million input tokens and $0.40 per million output tokens, according to Artificial Analysis. We track its price at 2 providers. At a mix of three input tokens to one output token, it is cheaper than 86% of the 330 priced models we track.

How many benchmarks has GPT-5 nano been tested on?

We track 85 results for GPT-5 nano on 43 benchmarks from 21 sources, 34 of them independently verified. The latest was recorded on Oct 8, 2026.

Which API features does GPT-5 nano support?

OpenRouter lists tool calling, structured outputs, and reasoning for GPT-5 nano.

About this record

Where GPT-5 nano's numbers come from, and every name it appears under.

Tracked since
May 2, 2026
Newest source mention
Aug 10, 2026

Where the results come from

Verification: 85 scores · 34 independently verified · 46 aggregator-attributed · 5 vendor-reported. How these tiers are assigned

From 21 sources on 11 sites. Artificial Analysis supplies 47 of them; the 34 independently verified results come from 10 sites. Bars are coloured by trust tier.

  • artificialanalysis.ai47
  • arcprize.org7
  • epoch.ai6
  • storage.googleapis.com6
  • api.llm-stats.com5
  • huggingface.co4
  • raw.githubusercontent.com4
  • matharena.ai2
  • swebench.com2
  • datasets-server.huggingface.co1
  • lmarena.ai1

Also known as

How our sources name GPT-5 nano at each reasoning setting.

SettingShort formAPI id
minimalgpt-5 nano (minimal)gpt-5-nano-minimal
lowgpt-5 nano (low)—
mediumgpt-5 nano (2025-08-07) (medium reasoning) gpt-5 nano (medium) gpt 5 nano (2025-08-07) (medium)gpt-5-nano-medium
highgpt-5 nano (high) GPT-5 nano (high) (Non-Reasoning) gpt-5 nano (high) (reasoning)gpt-5-nano-high
Also listed asgpt-5 nano (2025-08-07)gpt-5-nano (inspect)gpt-5-nano-2025-08-07gpt-5-nano:batch