Models / GPT-4.1

OpenAI

GPT-4.1

GPT-4.1 is strongest at agents and tool use; most buyers will do better with one of the alternatives below. The cheapest model of similar intelligence is gpt-oss-20b, at $0.06 per million tokens against $3.50.

Superseded. GPT-5.5 is OpenAI's newer model in this line, at $11.25 blended: see GPT-5.5 or compare the two.

Retiring 14 Jan 2027. OpenAI retires gpt-4.1; the official replacement is gpt-6-luna. How to re-test before you switch.

Results 16 Nov 2026 · 36 results · 6 sources

Test GPT-4.1 on your prompt → It runs beside the models you could use instead, with every answer checked and costed.

Price per million tokens
$2.00 in · $8.00 out
Typical host on OpenRouter, 10 Nov 2026 · compare prices
Speed
Not measured
Intelligence Index
12.7
Artificial Analysis
Spring Prompt overall
Not ranked yet
needs our benchmarks and two groups of results
Developer
OpenAI, based in the US
Weights not published
Released
14 Apr 2025
1,047,576 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 124th of 185Claude Opus 5.5 – $3.50 blended, 82nd percentilecheapest: Mistral Nemo –
Models priced $3–10 per million tokens 33rd of 38Claude Opus 5.5 – sets this group –
Models with similar intelligence sets this group – $3.50 blended, 100th percentilecheapest: gpt-oss-20b –
Models from developers based in the US 67th of 92Claude Opus 5.5 – $3.50 blended, 66th 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 $3–10 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 46th of 159 available · 20th of 35 at its price
BenchmarkGPT-4.1Rank among availableBest available
Berkeley Function Calling Leaderboard (BFCL) V4BFCL: overall accuracy · reported 54.0% 11th 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 143rd of 187 available · 34th of 39 at its price
BenchmarkGPT-4.1Rank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 12.7 124th of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 4.2% 165th of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 66.6% 124th of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 57th of 169 available · 24th of 34 at its price
BenchmarkGPT-4.1Rank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,415 86th of 153 Gemini 4 Argon 1,525
UGI LeaderboardUGI: writing score · reported 67.5 17th of 124 Gemini 3.8 Flash 78.6
Sticking to the factsSummarising without inventing things, and factual answers 87th of 139 available · 24th of 31 at its price
BenchmarkGPT-4.1Rank among availableBest available
Vectara Hallucination LeaderboardVectara: hallucination rate · reported 5.6% 14th of 71 GPT-5.4 nano 3.1%
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,408 93rd of 109 Gemini 4 Argon 1,511
SimpleQA Verified (Epoch AI)SimpleQA Verified: correct answers · reported 31.1% 53rd of 68 GPT-6 Astra 75.6%
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

Against GPT-4.1 at $3.50 blended, Intelligence Index 12.7. "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 $3.50 blended per million tokens. Cheapest first.

HostQuantisationInputOutputBlended
Azurenot stated$2.00$8.00$3.50
OpenAInot stated$2.00$8.00$3.50
Azurenot stated$2.20$8.80$3.85

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-4.1Claude Opus 5.5
most intelligent at $3–10 per million tokens
gpt-oss-20b
cheapest with similar intelligence
Claude Fable 5.2
leads overall
GPT-6.1 Sol
OpenAI's best other model
Price per million tokens $3.50 $8.00$0.06$16.00$4.00
Tokens a second – 96277–53
Intelligence Index 12.7 57.610.0–51.8
Spring Prompt overall – 82–9080
Rank among available models, by use case
Product listings – 13th–5th2nd
Decks from an analysis – 7th–4th5th
User surveys – 4th–1st3rd
Marketing planning – 9th–1st–
Agents and tool use 46th of 159 3rd130th–6th
Professional work – 3rd––9th
Reasoning and knowledge 143rd of 187 1st132nd–6th
Writing 57th of 169 5th155th–16th
Sticking to the facts 87th of 139 3rd––18th
Speed under pressure – ––8th3rd

Shaded figures are better than GPT-4.1's. A dash means no result.

Where it has been measured

33rd most intelligent of the 38 models priced $3–10 per million tokens. Among models available today, in the bottom quarter for reasoning and knowledge.

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

Every result

36 published results from 6 sources, each in the source's own units, with the configuration that produced it.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 7 results
MeasureValueRankConfigurationDated
Arena Text factuality: overallArena rating, higher is better 1,408
1,402–1,415
147 of 182 GPT-4.1 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,415
1,408–1,422
132 of 406 GPT-4.1 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,403
1,395–1,410
105 of 411 GPT-4.1 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,405
1,393–1,417
168 of 363 GPT-4.1 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,403
1,398–1,409
137 of 413 GPT-4.1 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,415
1,411–1,418
139 of 413 GPT-4.1 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,405
1,399–1,412
115 of 412 GPT-4.1 2 Oct 2026

Not shown: Not an error rate: claims that cannot be checked on the web are skipped, and preference still carries most of the weight.

Artificial Analysis · reported by Artificial Analysis · 7 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 12.7 283 of 413 GPT-4.1 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 66.6% 269 of 360 GPT-4.1 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 4.2% 374 of 419 GPT-4.1 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 43.0% 184 of 288 GPT-4.1 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 68.3% 213 of 407 GPT-4.1 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 13.6% 182 of 282 GPT-4.1 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 47.1% 166 of 286 GPT-4.1 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 86.5% 27 of 109 GPT-4.1 (bfcl fc) 16 Dec 2025
BFCL: irrelevance detection% correct, higher is better 84.0% 38 of 109 GPT-4.1 (bfcl prompt) 16 Dec 2025
BFCL: memory% correct, higher is better 23.9% 35 of 109 GPT-4.1 (bfcl fc) 16 Dec 2025
BFCL: memory% correct, higher is better 21.5% 41 of 109 GPT-4.1 (bfcl prompt) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 38.9% 29 of 109 GPT-4.1 (bfcl fc) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 9.8% 70 of 109 GPT-4.1 (bfcl prompt) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 54.0% 20 of 109 GPT-4.1 (bfcl fc) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 39.4% 45 of 109 GPT-4.1 (bfcl prompt) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 87.5% 27 of 109 GPT-4.1 (bfcl fc) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 100.0% 1 of 109 GPT-4.1 (bfcl prompt) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 82.8% 60 of 109 GPT-4.1 (bfcl fc) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 88.7% 12 of 109 GPT-4.1 (bfcl prompt) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 70.0% 59 of 109 GPT-4.1 (bfcl fc) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 78.9% 18 of 109 GPT-4.1 (bfcl prompt) 16 Dec 2025
BFCL: web search% correct, higher is better 68.0% 18 of 109 GPT-4.1 (bfcl fc) 16 Dec 2025
BFCL: web search% correct, higher is better 35.0% 32 of 109 GPT-4.1 (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 31.1% 63 of 85 GPT-4.1 31 Aug 2026

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

UGI Leaderboard · reported by UGI Leaderboard · 3 results
MeasureValueRankConfigurationDated
UGI: requested-length error% off the requested word count, lower is better 6.0% 49 of 379 GPT-4.1 24 Sep 2025
UGI: style adherencescore from 0 to 1, higher is better 0.33 275 of 379 GPT-4.1 24 Sep 2025
UGI: writing scorescore out of 100, higher is better 67.5 45 of 379 GPT-4.1 24 Sep 2025

Not shown: Not other format limits such as character counts or bullet counts.

Vectara Hallucination Leaderboard · reported by Vectara Hallucination Leaderboard · 2 results
MeasureValueRankConfigurationDated
Vectara: answer rate% of documents, higher is better 99.9% 19 of 108 GPT-4.1 22 Sep 2026
Vectara: hallucination rate% of summaries, lower is better 5.6% 21 of 108 GPT-4.1 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-4.1 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.