Models / GPT-5 mini

OpenAI

GPT-5 mini

GPT-5 mini trails most of the field on what it has been measured on; worth a look for simple, high-volume work where cost matters most. The cheapest model of similar intelligence is Gemma 4 26B A4B, at $0.14 per million tokens against $0.69.

Superseded. GPT-5.4 mini is OpenAI's newer model in this line, at $1.69 blended: see GPT-5.4 mini or compare the two.

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

Results 16 Nov 2026 · 53 results · 6 sources

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

Price per million tokens
$0.25 in · $2.00 out
Typical host on OpenRouter, 10 Nov 2026 · compare prices
Speed
Not measured
Intelligence Index
20.6
Artificial Analysis, at medium reasoning; 9.9–20.6 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 98th of 185Claude Opus 5.5 – $0.69 blended, 46th percentilecheapest: Mistral Nemo –
Models priced $0.25–1 per million tokens 23rd of 61MiMo-V2.6-Pro – sets this group –
Models with similar intelligence sets this group – $0.69 blended, 35th percentilecheapest: Gemma 4 26B A4B –
Models from developers based in the US 55th of 92Claude Opus 5.5 – $0.69 blended, 33rd 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 $0.25–1 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 112th of 159 available · 29th of 47 at its price
BenchmarkGPT-5 miniRank among availableBest available
Berkeley Function Calling Leaderboard (BFCL) V4BFCL: overall accuracy · reported 55.5% 10th of 36 Claude Opus 4.5 77.5%
Artificial AnalysisArtificial Analysis: Terminal-Bench 4.0 · reported 0.0% 67th of 107 Kimi K3.1 Pro 67.2%
Artificial AnalysisArtificial Analysis: Terminal-Bench 2.1 · reported 3.7% 101st of 107 Claude Fable 5.1 91.4%
Artificial AnalysisArtificial Analysis: τ-bench banking · reported 15.5% 53rd of 104 Grok 4.6 50.7%
Vending-Bench 2Vending-Bench 2: money after a year · reported $-31.18 53rd of 53 GPT-6 Astra $15,515
  • tau2-bench: has not measured this model.
  • OpenHands Index: has not measured this model.
  • Microsoft STATE-Bench: 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 101st of 187 available · 26th of 61 at its price
BenchmarkGPT-5 miniRank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 20.6 98th of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 21.5% 98th of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 82.8% 83rd of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 113th of 169 available · 38th of 58 at its price
BenchmarkGPT-5 miniRank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,390 104th of 153 Gemini 4 Argon 1,525
  • UGI Leaderboard: has not measured this model.
Sticking to the factsSummarising without inventing things, and factual answers 124th of 139 available · 34th of 40 at its price
BenchmarkGPT-5 miniRank among availableBest available
Vectara Hallucination LeaderboardVectara: hallucination rate · reported 12.9% 59th of 71 GPT-5.4 nano 3.1%
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,394 99th of 109 Gemini 4 Argon 1,511
SimpleQA Verified (Epoch AI)SimpleQA Verified: correct answers · reported 21.6% 58th 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-5 mini at $0.69 blended, Intelligence Index 20.6. "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 $0.69 blended per million tokens. Cheapest first.

HostQuantisationInputOutputBlended
Azurenot stated$0.25$2.00$0.69
OpenAInot stated$0.25$2.00$0.69
Azurenot stated$0.28$2.20$0.76

Other service tiers, not counted above: OpenAI flex $0.12 in, $1.00 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 miniMiMo-V2.6-Pro
most intelligent at $0.25–1 per million tokens
Gemma 4 26B A4B
cheapest with similar intelligence
Claude Opus 5.5
most intelligent available today
Price per million tokens $0.69 $0.54$0.14$8.00
Tokens a second – 438296
Intelligence Index 20.6 46.316.757.6
Spring Prompt overall – ––82
Rank among available models, by use case
Product listings – ––13th
Decks from an analysis – ––7th
User surveys – ––4th
Marketing planning – ––9th
Agents and tool use 112th of 159 23rd105th3rd
Professional work – ––3rd
Reasoning and knowledge 101st of 187 13th110th1st
Writing 113th of 169 23rd82nd5th
Sticking to the facts 124th of 139 10th43rd3rd
Speed under pressure – –––

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

Where it has been measured

23rd most intelligent of the 61 models priced $0.25–1 per million tokens. Among models available today, in the bottom quarter for sticking to the facts.

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

Every result

53 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 1,394
1,387–1,401
157 of 182 GPT-5 mini (high reasoning) 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,381
1,372–1,390
179 of 406 GPT-5 mini (high reasoning) 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,324
1,313–1,334
205 of 411 GPT-5 mini (high reasoning) 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,401
1,383–1,418
175 of 363 GPT-5 mini (high reasoning) 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,373
1,366–1,381
182 of 413 GPT-5 mini (high reasoning) 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,390
1,385–1,394
174 of 413 GPT-5 mini (high reasoning) 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,352
1,344–1,360
192 of 412 GPT-5 mini (high reasoning) 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 · 26 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Coding Indexindex score, higher is better 15.6 164 of 187 GPT-5 mini (high reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 16.8 237 of 413 GPT-5 mini (high reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 20.6 205 of 413 GPT-5 mini (medium reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 9.9 320 of 413 GPT-5 mini (minimal reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 82.8% 151 of 360 GPT-5 mini (high reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 80.3% 173 of 360 GPT-5 mini (medium reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 68.7% 252 of 360 GPT-5 mini (minimal reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 21.5% 193 of 419 GPT-5 mini (high reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 15.9% 224 of 419 GPT-5 mini (medium reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 5.1% 342 of 419 GPT-5 mini (minimal reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 75.4% 28 of 288 GPT-5 mini (high reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 71.2% 55 of 288 GPT-5 mini (medium reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 45.6% 163 of 288 GPT-5 mini (minimal reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 72.3% 178 of 407 GPT-5 mini (high reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 72.3% 178 of 407 GPT-5 mini (medium reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 39.3% 320 of 407 GPT-5 mini (minimal reasoning) 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 39.0% 168 of 205 GPT-5 mini (high reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 2.1% of tasks, higher is better 3.7% 174 of 188 GPT-5 mini (high reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 0.0% 152 of 204 GPT-5 mini (high reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 33.3% 88 of 282 GPT-5 mini (high reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 28.8% 112 of 282 GPT-5 mini (medium reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 14.4% 177 of 282 GPT-5 mini (minimal reasoning) 20 Oct 2026
Artificial Analysis: τ-bench banking% of tasks, higher is better 15.5% 101 of 175 GPT-5 mini (high reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 68.4% 132 of 286 GPT-5 mini (high reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 71.1% 125 of 286 GPT-5 mini (medium reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 31.9% 204 of 286 GPT-5 mini (minimal reasoning) 20 Oct 2026

Not shown: Not your codebase or tools.

Berkeley Function Calling Leaderboard (BFCL) V4 · reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 16 results
MeasureValueRankConfigurationDated
BFCL: irrelevance detection% correct, higher is better 91.0% 12 of 109 GPT-5 mini (bfcl fc) 16 Dec 2025
BFCL: irrelevance detection% correct, higher is better 55.7% 95 of 109 GPT-5 mini (bfcl prompt) 16 Dec 2025
BFCL: memory% correct, higher is better 44.3% 15 of 109 GPT-5 mini (bfcl fc) 16 Dec 2025
BFCL: memory% correct, higher is better 29.3% 22 of 109 GPT-5 mini (bfcl prompt) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 27.5% 43 of 109 GPT-5 mini (bfcl fc) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 5.5% 80 of 109 GPT-5 mini (bfcl prompt) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 55.5% 17 of 109 GPT-5 mini (bfcl fc) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 27.8% 77 of 109 GPT-5 mini (bfcl prompt) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 62.5% 91 of 109 GPT-5 mini (bfcl fc) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 93.8% 8 of 109 GPT-5 mini (bfcl prompt) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 69.9% 87 of 109 GPT-5 mini (bfcl fc) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 68.0% 90 of 109 GPT-5 mini (bfcl prompt) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 58.6% 90 of 109 GPT-5 mini (bfcl fc) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 62.6% 82 of 109 GPT-5 mini (bfcl prompt) 16 Dec 2025
BFCL: web search% correct, higher is better 82.0% 4 of 109 GPT-5 mini (bfcl fc) 16 Dec 2025
BFCL: web search% correct, higher is better 8.5% 58 of 109 GPT-5 mini (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 21.6% 70 of 85 GPT-5 mini (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 99.9% 19 of 108 GPT-5 mini 22 Sep 2026
Vectara: hallucination rate% of summaries, lower is better 12.9% 86 of 108 GPT-5 mini 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.

Vending-Bench 2 · reported by Vending-Bench 2 · 1 result
MeasureValueRankConfigurationDated
Vending-Bench 2: money after a yearUS dollars, higher is better $-31.18 63 of 63 GPT-5 mini 1 Oct 2026

Not shown: Not a real business; one simulated market with set rules.

Will GPT-5 mini 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.