Models / GPT-5 nano

OpenAI

GPT-5 nano

GPT-5 nano is strongest at agents and tool use; worth a look for simple, high-volume work where cost matters most. The cheapest model of similar intelligence is gpt-oss-20b, at $0.06 per million tokens against $0.14.

Superseded. GPT-5.4 nano is OpenAI's newer model in this line, at $0.46 blended: see GPT-5.4 nano or compare the two.

Retiring 11 Dec 2026. OpenAI retires gpt-5-nano; the official replacement is gpt-5.6-luna. How to re-test before you switch.

Results 16 Nov 2026 · 46 results · 5 sources

Test GPT-5 nano on your prompt → It runs beside the models you could use instead, with every answer checked and costed.

Price per million tokens
$0.05 in · $0.40 out
Typical host on OpenRouter, 10 Nov 2026 · compare prices
Speed
Not measured
Intelligence Index
13.0
Artificial Analysis, at high reasoning; 7.1–13.0 across 3 settings
Spring Prompt overall
Not ranked yet
needs our benchmarks and two groups of results
Developer
OpenAI, based in the US
Weights not published
Released
7 Aug 2025
400,000 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 121st of 185Claude Opus 5.5 – $0.14 blended, 10th percentilecheapest: Mistral Nemo –
Models priced under $0.25 per million tokens 13th of 41Claude Haiku 5.5 – sets this group –
Models with similar intelligence sets this group – $0.14 blended, 10th percentilecheapest: gpt-oss-20b –
Models from developers based in the US 65th of 92Claude Opus 5.5 – $0.14 blended, 12th percentilecheapest: Gemma 3 4B –

The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced under $0.25 per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.

Product listingsTurning a sparse product feed and photos into listings that can go live Not measured
  • CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is Not measured
  • DeckBench: has not measured this model.
User surveysPlanning a user survey and reading its results without being misled Not measured
  • SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending Not measured
  • ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 57th of 159 available · 6th of 37 at its price
BenchmarkGPT-5 nanoRank among availableBest available
Berkeley Function Calling Leaderboard (BFCL) V4BFCL: overall accuracy · reported 51.5% 13th of 36 Claude Opus 4.5 77.5%
  • tau2-bench: has not measured this model.
  • OpenHands Index: has not measured this model.
  • Microsoft STATE-Bench: has not measured this model.
  • Artificial Analysis: has not measured this model.
  • Vending-Bench 2: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects Not measured
  • APEX-Agents: has not measured this model.
  • GDP.pdf: has not measured this model.
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 129th of 187 available · 14th of 41 at its price
BenchmarkGPT-5 nanoRank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 13.0 121st of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 9.5% 131st of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 67.6% 120th of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 133rd of 169 available · 19th of 39 at its price
BenchmarkGPT-5 nanoRank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,338 124th of 153 Gemini 4 Argon 1,525
  • UGI Leaderboard: has not measured this model.
Sticking to the factsSummarising without inventing things, and factual answers 116th of 139 available · 22nd of 30 at its price
BenchmarkGPT-5 nanoRank among availableBest available
Vectara Hallucination LeaderboardVectara: hallucination rate · reported 10.5% 43rd of 71 GPT-5.4 nano 3.1%
SimpleQA Verified (Epoch AI)SimpleQA Verified: correct answers · reported 11.7% 64th of 68 GPT-6 Astra 75.6%
  • Arena (formerly LMArena): has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

Against GPT-5 nano at $0.14 blended, Intelligence Index 13.0. "Similar intelligence" means within 4 points or better.

Where to run it

3 hosts on OpenRouter at the standard service tier, with a median of $0.14 blended per million tokens. Cheapest first.

HostQuantisationInputOutputBlended
Azurenot stated$0.05$0.40$0.14
OpenAInot stated$0.05$0.40$0.14
Azurenot stated$0.06$0.44$0.15

Other service tiers, not counted above: OpenAI flex $0.03 in, $0.20 out. Flex and batch are cheaper but slower or queued; priority and fast cost more.

Prices per million tokens from OpenRouter's public endpoint list, checked 10 Nov 2026.

Against the alternatives

The models you would most likely weigh it against: the leaders of the groups above.

Scroll sideways to see every alternative.

GPT-5 nanoClaude Haiku 5.5
most intelligent at under $0.25 per million tokens
gpt-oss-20b
cheapest with similar intelligence
Claude Opus 5.5
most intelligent available today
Price per million tokens $0.14 $0.20$0.06$8.00
Tokens a second – 24027796
Intelligence Index 13.0 43.410.057.6
Spring Prompt overall – 40–82
Rank among available models, by use case
Product listings – 20th–13th
Decks from an analysis – 18th–7th
User surveys – 7th–4th
Marketing planning – 30th–9th
Agents and tool use 57th of 159 28th130th3rd
Professional work – ––3rd
Reasoning and knowledge 129th of 187 25th132nd1st
Writing 133rd of 169 –155th5th
Sticking to the facts 116th of 139 ––3rd
Speed under pressure – 21st––

Shaded figures are better than GPT-5 nano's. A dash means no result.

Where it has been measured

13th most intelligent of the 41 models priced under $0.25 per million tokens. Among models available today, in the bottom quarter for writing and sticking to the facts.

Artificial Analysis lists it but has not published a speed for it yet

Every result

46 published results from 5 sources, each in the source's own units, with the configuration that produced it.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 6 results
MeasureValueRankConfigurationDated
Arena Text: business, management and financeArena rating, higher is better 1,339
1,323–1,354
229 of 406 GPT-5 nano (high reasoning) 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,247
1,227–1,267
294 of 411 GPT-5 nano (high reasoning) 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,354
1,321–1,386
216 of 363 GPT-5 nano (high reasoning) 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,327
1,314–1,340
226 of 413 GPT-5 nano (high reasoning) 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,338
1,331–1,344
233 of 413 GPT-5 nano (high reasoning) 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,289
1,275–1,303
273 of 412 GPT-5 nano (high reasoning) 2 Oct 2026

Not shown: Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.

Artificial Analysis · reported by Artificial Analysis · 21 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 13.0 280 of 413 GPT-5 nano (high reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 12.5 287 of 413 GPT-5 nano (medium reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 7.1 371 of 413 GPT-5 nano (minimal reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 67.6% 259 of 360 GPT-5 nano (high reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 67.0% 265 of 360 GPT-5 nano (medium reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 42.8% 339 of 360 GPT-5 nano (minimal reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 9.5% 287 of 419 GPT-5 nano (high reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 8.7% 292 of 419 GPT-5 nano (medium reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 4.0% 384 of 419 GPT-5 nano (minimal reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 67.6% 77 of 288 GPT-5 nano (high reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 65.9% 85 of 288 GPT-5 nano (medium reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 32.5% 253 of 288 GPT-5 nano (minimal reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 45.0% 297 of 407 GPT-5 nano (high reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 43.7% 303 of 407 GPT-5 nano (medium reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 20.0% 364 of 407 GPT-5 nano (minimal reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 12.1% 190 of 282 GPT-5 nano (high reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 17.4% 161 of 282 GPT-5 nano (medium reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 6.8% 215 of 282 GPT-5 nano (minimal reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 36.6% 186 of 286 GPT-5 nano (high reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 30.4% 212 of 286 GPT-5 nano (medium reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 25.7% 232 of 286 GPT-5 nano (minimal reasoning) 20 Oct 2026

Not shown: Not business work, and a blend: read the parts for any one task.

Berkeley Function Calling Leaderboard (BFCL) V4 · reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 16 results
MeasureValueRankConfigurationDated
BFCL: irrelevance detection% correct, higher is better 89.1% 16 of 109 GPT-5 nano (bfcl fc) 16 Dec 2025
BFCL: irrelevance detection% correct, higher is better 45.8% 100 of 109 GPT-5 nano (bfcl prompt) 16 Dec 2025
BFCL: memory% correct, higher is better 24.7% 31 of 109 GPT-5 nano (bfcl fc) 16 Dec 2025
BFCL: memory% correct, higher is better 24.5% 33 of 109 GPT-5 nano (bfcl prompt) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 34.5% 36 of 109 GPT-5 nano (bfcl fc) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 0.8% 101 of 109 GPT-5 nano (bfcl prompt) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 51.5% 24 of 109 GPT-5 nano (bfcl fc) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 27.6% 79 of 109 GPT-5 nano (bfcl prompt) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 75.0% 62 of 109 GPT-5 nano (bfcl fc) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 93.8% 8 of 109 GPT-5 nano (bfcl prompt) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 68.0% 91 of 109 GPT-5 nano (bfcl fc) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 80.8% 75 of 109 GPT-5 nano (bfcl prompt) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 59.4% 88 of 109 GPT-5 nano (bfcl fc) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 70.7% 56 of 109 GPT-5 nano (bfcl prompt) 16 Dec 2025
BFCL: web search% correct, higher is better 72.5% 14 of 109 GPT-5 nano (bfcl fc) 16 Dec 2025
BFCL: web search% correct, higher is better 13.5% 50 of 109 GPT-5 nano (bfcl prompt) 16 Dec 2025

Not shown: Not general refusal or safety behaviour.

SimpleQA Verified (Epoch AI) · reported by SimpleQA Verified (Epoch AI) · 1 result
MeasureValueRankConfigurationDated
SimpleQA Verified: correct answers% of questions, higher is better 11.7% 81 of 85 GPT-5 nano (high reasoning) 10 Aug 2026

Not shown: Not answers grounded in your documents; tests what the model remembers.

Vectara Hallucination Leaderboard · reported by Vectara Hallucination Leaderboard · 2 results
MeasureValueRankConfigurationDated
Vectara: answer rate% of documents, higher is better 100.0% 1 of 108 GPT-5 nano 22 Sep 2026
Vectara: hallucination rate% of summaries, lower is better 10.5% 62 of 108 GPT-5 nano 22 Sep 2026

Not shown: Not a quality score: a low rate usually means content filters were triggered, and hallucination rates are measured on answered documents only.

Will GPT-5 nano work for your product?

Benchmarks aren't your task. Paste the prompt your product uses and run it beside the models you could use instead, with every answer checked and costed.

New here? Your email starts a free trial. No credit card needed.