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

GPT-5.4 nano

OpenAIReleased Mar 17, 2026Current

Score basis

GPT-5.4 nano is capable in long context and instruction following; and behind the leaders in reasoning, multimodal tasks, coding, factuality, agentic tasks, and math. Too few results yet to rate safety or multilingual tasks.

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.

Capable2 capabilities
  1. Long Context−16.7%1 of 3
  2. Instruction Following−23.8%2 of 3
Limited6 capabilities
  1. Reasoning−29.4%5 of 6
  2. Multimodal−33.9%1 of 6
  3. Coding−35.1%6 of 10
  4. Factuality−36.8%4 of 4
  5. Agentic−41.5%5 of 7
  6. Math−47.2%3 of 5
Not rated2 capabilities

Too few results yet to rate Safety or Multilingual.

Price

$0.20input$1.25outputper million tokens

From Artificial Analysis · 2 providers tracked · All prices

Cheaper than 60% of 330 priced models · 3:1 input-to-output blend, log scale

Evidence

92results on47benchmarks

  • 27 independently verified
  • 54 aggregator
  • 11 vendor-reported

From 11 sources · latest Oct 8, 2026 · How verification works

API features

Tool callingStructured outputsReasoning

As listed by OpenRouter

Research

9 papers reference GPT-5.4 nano

GPT-5.4 nano benchmark results

92 results on 47 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Long Context

CapableFull long context ranking

16.7% behind the leader1 of 3 ranked benchmarks measured

Show 2 more long context resultsHide 2 long context results

Instruction Following

CapableFull instruction following ranking

23.8% behind the leader2 of 3 ranked benchmarks measured

Show 2 more instruction following resultsHide 2 instruction following results

Reasoning

LimitedFull reasoning ranking

29.4% behind the leader5 of 6 ranked benchmarks measured

Show 11 more reasoning resultsHide 11 reasoning results

Multimodal

LimitedFull multimodal ranking

33.9% behind the leader1 of 6 ranked benchmarks measured

Show 5 more multimodal resultsHide 5 multimodal results

Coding

LimitedFull coding ranking

35.1% behind the leader6 of 10 ranked benchmarks measured

Show 4 more coding resultsHide 4 coding results

Factuality

LimitedFull factuality ranking

36.8% behind the leader4 of 4 ranked benchmarks measured

Show 4 more factuality resultsHide 4 factuality results

Agentic

LimitedFull agentic ranking

41.5% behind the leader5 of 7 ranked benchmarks measured

Show 5 more agentic resultsHide 5 agentic results

Math

LimitedFull math ranking

47.2% behind the leader3 of 5 ranked benchmarks measured

Show 2 more math resultsHide 2 math results

More results

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

Show 24 more resultsHide 24 results

GPT-5.4 nano: common questions

Who makes GPT-5.4 nano?

GPT-5.4 nano is made by OpenAI.

When was GPT-5.4 nano released?

GPT-5.4 nano was released on Mar 17, 2026, according to Artificial Analysis.

What is GPT-5.4 nano good at?

GPT-5.4 nano is capable in long context and instruction following; and behind the leaders in reasoning, multimodal tasks, coding, factuality, agentic tasks, and math. Too few results yet to rate safety or multilingual tasks.

How much does GPT-5.4 nano cost?

GPT-5.4 nano costs $0.20 per million input tokens and $1.25 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 60% of the 330 priced models we track.

How many benchmarks has GPT-5.4 nano been tested on?

We track 92 results for GPT-5.4 nano on 47 benchmarks from 11 sources, 27 of them independently verified. The latest was recorded on Oct 8, 2026.

Which API features does GPT-5.4 nano support?

OpenRouter lists tool calling, structured outputs, and reasoning for GPT-5.4 nano.

About this record

Where GPT-5.4 nano's numbers come from, and every name it appears under.

Tracked since
Apr 27, 2026
Newest source mention
Sep 9, 2026

Where the results come from

Verification: 92 scores · 27 independently verified · 54 aggregator-attributed · 11 vendor-reported. How these tiers are assigned

From 11 sources on 8 sites. Artificial Analysis supplies 54 of them; the 27 independently verified results come from 6 sites. Bars are coloured by trust tier.

  • artificialanalysis.ai54
  • api.llm-stats.com11
  • arcprize.org8
  • livebench.ai7
  • epoch.ai6
  • raw.githubusercontent.com4
  • datasets-server.huggingface.co1
  • lmarena.ai1

Also known as

How our sources name GPT-5.4 nano at each reasoning setting.

SettingShort formAPI id
lowgpt-5.4 nano (low)—
mediumgpt-5.4 nano (medium)gpt-5-4-nano-medium
highgpt-5.4 nano (high)gpt-5.4-nano-high
xhighgpt-5.4 nano (xhigh)—
Also listed asgpt-5.4-nano-2026-03-17gpt-5.4 nano (non-reasoning)gpt-5.4-nano (reasoning)gpt-5-4-nano-non-reasoninggpt-5-4-nanogpt-5.4-nano:batchgpt-5.4 nano (no thinking)