Models / GPT-5.4 mini
GPT-5.4 mini
GPT-5.4 mini is strongest at writing; worth a look for simple, high-volume work where cost matters most. The cheapest model of similar intelligence is DeepSeek-V4-Flash (0423), at $0.15 per million tokens against $1.69.
Results 16 Nov 2026 · 46 results · 5 sources
Test GPT-5.4 mini on your prompt → It runs beside the models you could use instead, with every answer checked and costed.
- Price per million tokens
- $0.75 in · $4.50 out
OpenAI's list price, 10 Nov 2026 · compare prices$0.07 cached input - Speed
- Not measured
- Intelligence Index
- 24.1
Artificial Analysis, at xhigh reasoning; 11.1–24.1 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
- 17 Mar 2026
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.
| Among | Intelligence Index | Spring Prompt overall | Price, blended | Speed |
|---|---|---|---|---|
| Models available today | 78th of 185Claude Opus 5.5 | – | $1.69 blended, 64th percentilecheapest: Mistral Nemo | – |
| Models priced $1–3 per million tokens | 22nd of 34Muse Spark 1.3 | – | sets this group | – |
| Models with similar intelligence | sets this group | – | $1.69 blended, 68th percentilecheapest: DeepSeek-V4-Flash (0423) | – |
| Models from developers based in the US | 47th of 92Claude Opus 5.5 | – | $1.69 blended, 46th 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.
Measured cost per task
API cost recorded in our benchmark runs, excluding judging.
| Task | Benchmark | Cost per task |
|---|---|---|
| Game of bullet chess | BulletBench | ≈ $0.021 |
How good is it, and for what?
Each use case ranks the available models its benchmarks measured, then those priced $1–3 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
- CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is
- DeckBench: has not measured this model.
User surveysPlanning a user survey and reading its results without being misled
- SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending
- ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls
| Benchmark | GPT-5.4 mini | Rank among available | Best available |
|---|---|---|---|
| Artificial AnalysisArtificial Analysis: Terminal-Bench 4.0 · reported | 2.0% | 52nd of 107 | Kimi K3.1 Pro 67.2% |
| Artificial AnalysisArtificial Analysis: Terminal-Bench 2.1 · reported | 59.2% | 49th of 107 | Claude Fable 5.1 91.4% |
| Artificial AnalysisArtificial Analysis: τ-bench banking · reported | 25.6% | 37th of 104 | Grok 4.6 50.7% |
- tau2-bench: has not measured this model.
- Berkeley Function Calling Leaderboard (BFCL) V4: pending. Its latest results are dated 16 Dec 2025, before this model was released.
- OpenHands Index: has not measured this model.
- Microsoft STATE-Bench: 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
- 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
| Benchmark | GPT-5.4 mini | Rank among available | Best available |
|---|---|---|---|
| Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported | 24.1 | 78th of 185 | Claude Opus 5.5 57.6 |
| Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported | 28.1% | 83rd of 186 | Claude Opus 5.5 61.4% |
| Artificial AnalysisArtificial Analysis: GPQA Diamond · reported | 87.5% | 55th of 168 | GPT-6 Astra 96.3% |
WritingWhat people prefer in blind comparisons, and judged writing quality
| Benchmark | GPT-5.4 mini | Rank among available | Best available |
|---|---|---|---|
| Arena (formerly LMArena)Arena Text: overall · reported | 1,447 | 55th of 153 | Gemini 4 Argon 1,525 |
- UGI Leaderboard: has not measured this model.
Sticking to the factsSummarising without inventing things, and factual answers
| Benchmark | GPT-5.4 mini | Rank among available | Best available |
|---|---|---|---|
| Vectara Hallucination LeaderboardVectara: hallucination rate · reported | 5.5% | 11th of 71 | GPT-5.4 nano 3.1% |
| Arena (formerly LMArena)Arena Text factuality: overall · reported | 1,447 | 53rd of 109 | Gemini 4 Argon 1,511 |
| SimpleQA Verified (Epoch AI)SimpleQA Verified: correct answers · reported | 29.4% | 55th of 68 | GPT-6 Astra 75.6% |
Speed under pressureGood decisions against a real clock (fast chess)
| Benchmark | GPT-5.4 mini | Rank among available | Best available |
|---|---|---|---|
| BulletBenchBulletBench Lightning 10+1: ladder Elo | 571 | 5th of 18 | Gemini 3.5 Flash-Lite 892 |
| BulletBenchBulletBench Bullet 60s: ladder Elo | 197 | 24th of 33 | Gemini 3.8 Flash 1,068 |
Nearest alternatives
- Cheaper at similar intelligenceDeepSeek-V4-Flash (0423)$0.15 blended · Intelligence Index 24.4Compare with GPT-5.4 mini →
- OpenAI's newer, at least as capable modelGPT-6 Astra$20.00 blended · Intelligence Index 52.7Compare with GPT-5.4 mini →
- Open weights at similar intelligenceDeepSeek-V4-Flash (0423)$0.15 blended · Intelligence Index 24.4Compare with GPT-5.4 mini →
Against GPT-5.4 mini at $1.69 blended, Intelligence Index 24.1. "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 $1.69 blended per million tokens. Cheapest first.
| Host | Quantisation | Input | Output | Blended |
|---|---|---|---|---|
| Azure | not stated | $0.75 | $4.50 | $1.69 |
| OpenAI | not stated | $0.75 | $4.50 | $1.69 |
| Azure | not stated | $0.82 | $4.95 | $1.86 |
Other service tiers, not counted above: OpenAI fast $1.50 in, $9.00 out; OpenAI flex $0.38 in, $2.25 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-5.4 mini | Muse Spark 1.3 most intelligent at $1–3 per million tokens | DeepSeek-V4-Flash (0423) cheapest with similar intelligence | Claude Opus 5.5 most intelligent available today | |
|---|---|---|---|---|
| Price per million tokens | $1.69 | $2.00 | $0.15 | $8.00 |
| Tokens a second | – | 254 | – | 96 |
| Intelligence Index | 24.1 | 48.1 | 24.4 | 57.6 |
| Spring Prompt overall | – | 59 | – | 82 |
| Rank among available models, by use case | ||||
| Product listings | – | 17th | – | 13th |
| Decks from an analysis | – | 25th | – | 7th |
| User surveys | – | – | – | 4th |
| Marketing planning | – | 25th | – | 9th |
| Agents and tool use | 64th of 159 | 13th | 59th | 3rd |
| Professional work | – | 8th | – | 3rd |
| Reasoning and knowledge | 79th of 187 | 11th | 69th | 1st |
| Writing | 59th of 169 | 7th | 63rd | 5th |
| Sticking to the facts | 63rd of 139 | 11th | 81st | 3rd |
| Speed under pressure | 15th of 35 | 33rd | – | – |
| Side by side → | Side by side → | Side by side → | ||
Shaded figures are better than GPT-5.4 mini's. A dash means no result.
Where it has been measured
22nd most intelligent of the 34 models priced $1–3 per million tokens. Among models available today, mid-field on every use case measured so far.
Artificial Analysis lists it but has not published a speed for it yet
- BulletBench10 results
- Arena (formerly LMArena)7 results
- Artificial Analysis26 results
- SimpleQA Verified (Epoch AI)1 result
- Vectara Hallucination Leaderboard2 results
- Berkeley Function Calling Leaderboard (BFCL) V4Pending: latest results 16 Dec 2025
- CatalogBenchNot measured
- DeckBenchNot measured
- ROASBenchNot measured
- SurveyBenchNot measured
- APEX-AgentsNot measured
- GDP.pdfNot measured
- Microsoft STATE-BenchNot measured
- OpenHands IndexNot measured
- Remote Labor IndexNot measured
- tau2-benchNot measured
- UGI LeaderboardNot measured
- Vending-Bench 2Not measured
Every result
46 published results from 5 sources, each in the source's own units, with the configuration that produced it.
BulletBench · measured by Spring Prompt · 10 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| BulletBench Bullet 60s: cost per gameUS dollars, lower is better | $0.0205 | 53 of 71 | GPT-5.4 mini (no reasoning) none reasoning |
1 Oct 2026 |
| BulletBench Bullet 60s: games lost on time% of games, lower is better | 50.0% | 41 of 71 | GPT-5.4 mini (no reasoning) none reasoning |
1 Oct 2026 |
| BulletBench Bullet 60s: invalid moves% of moves, lower is better | 0.0% | 1 of 71 | GPT-5.4 mini (no reasoning) none reasoning |
1 Oct 2026 |
| BulletBench Bullet 60s: ladder Eloladder Elo, higher is better | 197 0–445 |
56 of 71 | GPT-5.4 mini (no reasoning) none reasoning |
1 Oct 2026 |
| BulletBench Bullet 60s: median move timemilliseconds, lower is better | 0.9 s | 32 of 71 | GPT-5.4 mini (no reasoning) none reasoning |
1 Oct 2026 |
| BulletBench Lightning 10+1: cost per gameUS dollars, lower is better | $0.0136 | 41 of 48 | GPT-5.4 mini (no reasoning) none reasoning |
1 Oct 2026 |
| BulletBench Lightning 10+1: games lost on time% of games, lower is better | 0.0% | 1 of 48 | GPT-5.4 mini (no reasoning) none reasoning |
1 Oct 2026 |
| BulletBench Lightning 10+1: invalid moves% of moves, lower is better | 0.0% | 1 of 48 | GPT-5.4 mini (no reasoning) none reasoning |
1 Oct 2026 |
| BulletBench Lightning 10+1: ladder Eloladder Elo, higher is better | 571 412–721 |
15 of 48 | GPT-5.4 mini (no reasoning) none reasoning |
1 Oct 2026 |
| BulletBench Lightning 10+1: median move timemilliseconds, lower is better | 0.9 s | 35 of 48 | GPT-5.4 mini (no reasoning) none reasoning |
1 Oct 2026 |
Not shown: Not your cost: prices are those charged on the run date.
Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 7 results
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| Arena Text factuality: overallArena rating, higher is better | 1,447 1,444–1,450 |
74 of 182 | GPT-5.4 mini (high reasoning) | 2 Oct 2026 |
| Arena Text: business, management and financeArena rating, higher is better | 1,460 1,453–1,466 |
61 of 406 | GPT-5.4 mini (high reasoning) | 2 Oct 2026 |
| Arena Text: creative writingArena rating, higher is better | 1,401 1,394–1,409 |
107 of 411 | GPT-5.4 mini (high reasoning) | 2 Oct 2026 |
| Arena Text: expert promptsArena rating, higher is better | 1,480 1,472–1,489 |
72 of 363 | GPT-5.4 mini (high reasoning) | 2 Oct 2026 |
| Arena Text: instruction followingArena rating, higher is better | 1,434 1,428–1,439 |
89 of 413 | GPT-5.4 mini (high reasoning) | 2 Oct 2026 |
| Arena Text: overallArena rating, higher is better | 1,447 1,443–1,451 |
84 of 413 | GPT-5.4 mini (high reasoning) | 2 Oct 2026 |
| Arena Text: writing, literature and languageArena rating, higher is better | 1,423 1,417–1,429 |
88 of 412 | GPT-5.4 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
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| Artificial Analysis: Artificial Analysis Coding Indexindex score, higher is better | 56.1 | 85 of 187 | GPT-5.4 mini (xhigh reasoning) | 20 Oct 2026 |
| Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better | 19.7 | 214 of 413 | GPT-5.4 mini (medium reasoning) | 20 Oct 2026 |
| Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better | 11.1 | 306 of 413 | GPT-5.4 mini (no reasoning) | 20 Oct 2026 |
| Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better | 24.1 | 167 of 413 | GPT-5.4 mini (xhigh reasoning) | 20 Oct 2026 |
| Artificial Analysis: GPQA Diamond% of questions, higher is better | 82.3% | 156 of 360 | GPT-5.4 mini (medium reasoning) | 20 Oct 2026 |
| Artificial Analysis: GPQA Diamond% of questions, higher is better | 60.6% | 291 of 360 | GPT-5.4 mini (no reasoning) | 20 Oct 2026 |
| Artificial Analysis: GPQA Diamond% of questions, higher is better | 87.5% | 98 of 360 | GPT-5.4 mini (xhigh reasoning) | 20 Oct 2026 |
| Artificial Analysis: Humanity's Last Exam% of questions, higher is better | 18.6% | 213 of 419 | GPT-5.4 mini (medium reasoning) | 20 Oct 2026 |
| Artificial Analysis: Humanity's Last Exam% of questions, higher is better | 5.9% | 332 of 419 | GPT-5.4 mini (no reasoning) | 20 Oct 2026 |
| Artificial Analysis: Humanity's Last Exam% of questions, higher is better | 28.1% | 161 of 419 | GPT-5.4 mini (xhigh reasoning) | 20 Oct 2026 |
| Artificial Analysis: IFBench% of instructions, higher is better | 64.8% | 89 of 288 | GPT-5.4 mini (medium reasoning) | 20 Oct 2026 |
| Artificial Analysis: IFBench% of instructions, higher is better | 38.8% | 214 of 288 | GPT-5.4 mini (no reasoning) | 20 Oct 2026 |
| Artificial Analysis: IFBench% of instructions, higher is better | 73.3% | 38 of 288 | GPT-5.4 mini (xhigh reasoning) | 20 Oct 2026 |
| Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better | 67.0% | 222 of 407 | GPT-5.4 mini (medium reasoning) | 20 Oct 2026 |
| Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better | 37.0% | 324 of 407 | GPT-5.4 mini (no reasoning) | 20 Oct 2026 |
| Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better | 77.0% | 135 of 407 | GPT-5.4 mini (xhigh reasoning) | 20 Oct 2026 |
| Artificial Analysis: SciCode% of problems, higher is better | 52.1% | 88 of 205 | GPT-5.4 mini (xhigh reasoning) | 20 Oct 2026 |
| Artificial Analysis: Terminal-Bench 2.1% of tasks, higher is better | 59.2% | 92 of 188 | GPT-5.4 mini (xhigh reasoning) | 20 Oct 2026 |
| Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better | 2.0% | 125 of 204 | GPT-5.4 mini (xhigh reasoning) | 20 Oct 2026 |
| Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better | 34.1% | 84 of 282 | GPT-5.4 mini (medium reasoning) | 20 Oct 2026 |
| Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better | 18.2% | 155 of 282 | GPT-5.4 mini (no reasoning) | 20 Oct 2026 |
| Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better | 52.3% | 19 of 282 | GPT-5.4 mini (xhigh reasoning) | 20 Oct 2026 |
| Artificial Analysis: τ-bench banking% of tasks, higher is better | 25.6% | 73 of 175 | GPT-5.4 mini (xhigh reasoning) | 20 Oct 2026 |
| Artificial Analysis: τ²-bench telecom% of tasks, higher is better | 36.6% | 186 of 286 | GPT-5.4 mini (medium reasoning) | 20 Oct 2026 |
| Artificial Analysis: τ²-bench telecom% of tasks, higher is better | 23.4% | 242 of 286 | GPT-5.4 mini (no reasoning) | 20 Oct 2026 |
| Artificial Analysis: τ²-bench telecom% of tasks, higher is better | 83.3% | 93 of 286 | GPT-5.4 mini (xhigh reasoning) | 20 Oct 2026 |
Not shown: Not your codebase or tools.
SimpleQA Verified (Epoch AI) · reported by SimpleQA Verified (Epoch AI) · 1 result
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| SimpleQA Verified: correct answers% of questions, higher is better | 29.4% | 66 of 85 | GPT-5.4 mini (high reasoning) | 27 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
| Measure | Value | Rank | Configuration | Dated |
|---|---|---|---|---|
| Vectara: answer rate% of documents, higher is better | 100.0% | 1 of 108 | GPT-5.4 mini | 22 Sep 2026 |
| Vectara: hallucination rate% of summaries, lower is better | 5.5% | 18 of 108 | GPT-5.4 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.
Embed a score badge
A badge for each of our benchmarks that ranked GPT-5.4 mini, from the current results. Paste the code into a page or README; it links back to the benchmark.
BulletBench · Lightning 10+1
<a href="https://springprompt.com/benchmarks/bulletbench?focus=openai-gpt-5-4-mini"><img src="https://springprompt.com/badge/bulletbench/openai-gpt-5-4-mini.svg" alt="GPT-5.4 mini is #15 of 45 on BulletBench (Lightning 10+1: 571), measured by Spring Prompt" height="20"></a>
Will GPT-5.4 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.