GPT-5.4 Mini vs MiniMax M3
Updated October 21, 2026. 78 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick MiniMax M3 for coding: it scores 57.8 on the Vibe Coding Index against 28.2 for GPT-5.4 Mini.
Pick GPT-5.4 Mini when what matters is Terminal-Bench 4 (Vals AI) or Vibe Code Bench (Vals AI), where it leads.
- Shared results
- 78
- GPT-5.4 Mini only
- 10
- MiniMax M3 only
- 38
- Like-for-like categories
- 7 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, four public coding benchmarks
MiniMax M3
MiniMax M3 scores 57.8 against 28.2.
1 and 2 benchmarks behind the scores
Intelligence
AA Intelligence Index
MiniMax M3
MiniMax M3 leads, 29.2 against 24.1.
Same board, both measured
Agentic work
AA Agentic Index
MiniMax M3
MiniMax M3 leads, 29.5 against 17.9.
Same board, both measured
Reasoning
GPQA Diamond
MiniMax M3
MiniMax M3 leads, 92.7% against 83.1%.
Same board, both measured
Math
AIME
No call
Only GPT-5.4 Mini has a AIME result, 95.6%.
Directional only
Knowledge
MMLU Pro
GPT-5.4 Mini
GPT-5.4 Mini leads, 84.5% against 84.2%.
Same board, both measured
Web design
Design Arena: websites
No call
Only MiniMax M3 has a Design Arena: websites result, 1257.
Directional only
Long documents
Listed context window
MiniMax M3
MiniMax M3 takes 1.05M tokens against 400K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
MiniMax M3
GPT-5.4 Mini $0.0030, MiniMax M3 $0.0009 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
MiniMax M3
GPT-5.4 Mini $0.70, MiniMax M3 $0.56.
Published attempt cost and pass rate
At a glance
- Coding index
- GPT-5.4 Mini 28.2 from 1 benchmark, MiniMax M3 57.8 from 2 benchmarks.
- Widest category
- Agentic work, 8.6 points apart on average over 6 shared benchmarks.
- Widest single gap
- ALE-Bench: GPT-5.4 Mini 1188.6 against MiniMax M3 640.0, 18.3 points apart on a 0 to 100 scale.
- Output price
- GPT-5.4 Mini $4.50 against MiniMax M3 $1.20 per million output tokens, 3.8x apart.
- Context window
- GPT-5.4 Mini takes 400K tokens, MiniMax M3 1.05M.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | GPT-5.4 Mini | Bars | MiniMax M3 | Ties |
|---|---|---|---|---|
| Coding | 7 | 9 | ||
| Intelligence | 1 | 1 | ||
| Agentic work | 0 | 6 | ||
| Reasoning | 3 | 9 | ||
| Math | 2 | 2 | ||
| Knowledge | 1 | 7 | ||
| General | 5 | 25 | ||
| All shared boards | 19 | 59 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index79
| Benchmark | GPT-5.4 Mini | MiniMax M3 | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 70.9% | Only MiniMax M3 measured |
| LMArena WebDev | 1397 | 1482 | MiniMax M3 |
| AA Coding Index | 56.1 | 58.6 | MiniMax M3 |
| LiveCodeBench (Vals AI) | 81.5% | 82.2% | MiniMax M3 |
| SWE-bench Verified (Vals AI) | 73.0% | 75.0% | MiniMax M3 |
| Terminal-Bench 4 (Vals AI) | 2.5% | 1.0% | GPT-5.4 Mini |
| Vibe Code Bench (Vals AI) | 48.0% | 47.6% | GPT-5.4 Mini |
| Code migration (Vals AI) | 12.9% | 19.9% | MiniMax M3 |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only GPT-5.4 Mini measured |
| Terminal-Bench 2.1 (Vals AI) | 54.7% | 53.6% | GPT-5.4 Mini |
| SWE-bench Multimodal (Vals AI) | Not measured | 33.3% | Only MiniMax M3 measured |
| Vibe Code Bench 1-100 (Vals AI) | Not measured | 9.2% | Only MiniMax M3 measured |
| SkillsBench (Vals AI) | Not measured | 51.5% | Only MiniMax M3 measured |
| CyberBench (Vals AI) | Not measured | 65.0% | Only MiniMax M3 measured |
| SciCode | 52.1% | 47.1% | GPT-5.4 Mini |
| Terminal-Bench Hard (Artificial Analysis run) | 52.3% | 42.4% | GPT-5.4 Mini |
| FrontierCode | 27.0% | 14.7% | GPT-5.4 Mini |
| WeirdML | 60.3% | Not measured | Only GPT-5.4 Mini measured |
| ALE-Bench | 1188.6 | 640.0 | GPT-5.4 Mini |
| LMArena Text: coding | 1438 | 1470 | MiniMax M3 |
| LMArena Text: software and IT | 1436 | 1466 | MiniMax M3 |
| LMArena WebDev: React | 1386 | 1478 | MiniMax M3 |
| LMArena WebDev: HTML | 1419 | 1455 | MiniMax M3 |
| LMArena WebDev: from an image | Not measured | 1470 | Only MiniMax M3 measured |
| Neither model has a result on the other 14 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | GPT-5.4 Mini | MiniMax M3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 24.1 | 29.2 | MiniMax M3 |
| Epoch Capabilities Index | 148.7 | 146.9 | GPT-5.4 Mini |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index06
| Benchmark | GPT-5.4 Mini | MiniMax M3 | Leader |
|---|---|---|---|
| AA Agentic Index | 17.9 | 29.5 | MiniMax M3 |
| Design Arena: agents building web apps | Not measured | 1200 | Only MiniMax M3 measured |
| Design Arena: agents building full-stack apps | Not measured | 1197 | Only MiniMax M3 measured |
| Design Arena: agents building mobile apps | Not measured | 1169 | Only MiniMax M3 measured |
| Design Arena: agents building Android apps | Not measured | 1164 | Only MiniMax M3 measured |
| Design Arena: agents building games | Not measured | 1158 | Only MiniMax M3 measured |
| Design Arena: HTML slides | Not measured | 1180 | Only MiniMax M3 measured |
| Design Arena: PowerPoint slides | Not measured | 1205 | Only MiniMax M3 measured |
| Design Arena: spreadsheets | Not measured | 1021 | Only MiniMax M3 measured |
| LiveBench research | Not measured | 68.0% | Only MiniMax M3 measured |
| LiveBench browser | Not measured | 8.0% | Only MiniMax M3 measured |
| LiveBench long agentic | Not measured | 12.4% | Only MiniMax M3 measured |
| Finance Agent (Vals AI) | 45.4% | 48.3% | MiniMax M3 |
| Tax Agent Bench (Vals AI) | 8.8% | 20.6% | MiniMax M3 |
| Legal Research Bench (Vals AI) | 12.5% | 29.8% | MiniMax M3 |
| Excel Modeling Benchmark (Vals AI) | 45.4% | 47.8% | MiniMax M3 |
| τ²-Bench Telecom | 83.3% | 88.9% | MiniMax M3 |
| APEX-Agents | Not measured | 37.7% | Only MiniMax M3 measured |
| OSWorld 2.0 | Not measured | 4.6% | Only MiniMax M3 measured |
| DeepResearch Bench | 36.3% | Not measured | Only GPT-5.4 Mini measured |
| Vending-Bench 2 | Not measured | $2,158 | Only MiniMax M3 measured |
| Neither model has a result on the other 18 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index39
| Benchmark | GPT-5.4 Mini | MiniMax M3 | Leader |
|---|---|---|---|
| GPQA Diamond | 83.1% | 92.7% | MiniMax M3 |
| LiveBench reasoning | Not measured | 74.5% | Only MiniMax M3 measured |
| MMMU Pro (Vals AI) | 79.3% | 81.2% | MiniMax M3 |
| GPQA Diamond (Artificial Analysis run) | 87.5% | 92.9% | MiniMax M3 |
| Humanity's Last Exam (Artificial Analysis run) | 28.1% | 39.0% | MiniMax M3 |
| GPQA Diamond (Epoch AI run) | 86.9% | 90.9% | MiniMax M3 |
| ARC-AGI | 63.7% | Not measured | Only GPT-5.4 Mini measured |
| ARC-AGI-2 | 18.9% | Not measured | Only GPT-5.4 Mini measured |
| SimpleBench | Not measured | 45.8% | Only MiniMax M3 measured |
| Chess puzzles | 23.0% | 13.0% | GPT-5.4 Mini |
| Mystery game puzzles | 11.0% | 8.0% | GPT-5.4 Mini |
| LMArena Text: hard prompts | 1424 | 1447 | MiniMax M3 |
| LMArena Text: expert | 1435 | 1461 | MiniMax M3 |
| LMArena Text: hard prompts in English | 1420 | 1455 | MiniMax M3 |
| LMArena Vision: diagrams | 1266 | 1266 | MiniMax M3 |
| LMArena Vision: homework | 1285 | 1268 | GPT-5.4 Mini |
| Neither model has a result on the other 5 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index22
| Benchmark | GPT-5.4 Mini | MiniMax M3 | Leader |
|---|---|---|---|
| AIME | 95.6% | Not measured | Only GPT-5.4 Mini measured |
| OTIS Mock AIME | 88.9% | 71.1% | GPT-5.4 Mini |
| FrontierMath, tiers 1 to 3 | 51.2% | Not measured | Only GPT-5.4 Mini measured |
| FrontierMath, tier 4 | 9.8% | Not measured | Only GPT-5.4 Mini measured |
| ProofBench | 21.0% | 18.0% | GPT-5.4 Mini |
| LMArena Text: math | 1418 | 1429 | MiniMax M3 |
| LMArena Text: mathematical work | 1429 | 1448 | MiniMax M3 |
| Neither model has a result on the other 6 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index17
| Benchmark | GPT-5.4 Mini | MiniMax M3 | Leader |
|---|---|---|---|
| MMLU Pro | 84.5% | 84.2% | GPT-5.4 Mini |
| LegalBench (Vals AI) | Not measured | 85.4% | Only MiniMax M3 measured |
| CaseLaw (Vals AI) | 51.7% | Not measured | Only GPT-5.4 Mini measured |
| CorpFin (Vals AI) | 60.9% | 68.1% | MiniMax M3 |
| TaxEval (Vals AI) | 71.2% | 72.7% | MiniMax M3 |
| MortgageTax (Vals AI) | 63.5% | 68.4% | MiniMax M3 |
| MedCode (Vals AI) | Not measured | 46.3% | Only MiniMax M3 measured |
| SimpleQA Verified | 29.4% | Not measured | Only GPT-5.4 Mini measured |
| LMArena Text: science | 1417 | 1452 | MiniMax M3 |
| LMArena Text: medicine and healthcare | 1395 | 1441 | MiniMax M3 |
| LMArena Text: legal and government | 1414 | 1437 | MiniMax M3 |
| LMArena Text: business and finance | 1418 | 1434 | MiniMax M3 |
| Neither model has a result on the other 5 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | GPT-5.4 Mini | MiniMax M3 | Leader |
|---|---|---|---|
| Design Arena: websites | Not measured | 1257 | Only MiniMax M3 measured |
| Design Arena: UI components | Not measured | 1246 | Only MiniMax M3 measured |
| Design Arena: game dev | Not measured | 1210 | Only MiniMax M3 measured |
| Design Arena: data viz | Not measured | 1235 | Only MiniMax M3 measured |
| Design Arena: 3D | Not measured | 1209 | Only MiniMax M3 measured |
| Design Arena: all code categories | Not measured | 1247 | Only MiniMax M3 measured |
| Design Arena: SVG | Not measured | 1161 | Only MiniMax M3 measured |
| Design Arena: ASCII art | Not measured | 1161 | Only MiniMax M3 measured |
| LiveBench design | Not measured | 66.2% | Only MiniMax M3 measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index525
| Benchmark | GPT-5.4 Mini | MiniMax M3 | Leader |
|---|---|---|---|
| LiveBench data analysis | Not measured | 76.2% | Only MiniMax M3 measured |
| Public Benefits Bench (Vals AI) | Not measured | 64.1% | Only MiniMax M3 measured |
| SAGE (Vals AI) | 50.8% | 50.6% | GPT-5.4 Mini |
| MedScribe (Vals AI) | Not measured | 87.3% | Only MiniMax M3 measured |
| Vals Multimodal Index | 54.2% | 60.0% | MiniMax M3 |
| IFBench | 73.3% | 82.9% | MiniMax M3 |
| Long context reasoning (AA-LCR) | 77.0% | 83.0% | MiniMax M3 |
| DTBench | 80.0% | 78.9% | GPT-5.4 Mini |
| LMCA | 40.8% | 33.7% | GPT-5.4 Mini |
| Surface Evolver Bench | Not measured | 55.0% | Only MiniMax M3 measured |
| LMArena Text | 1412 | 1433 | MiniMax M3 |
| LMArena Text: instruction following | 1405 | 1433 | MiniMax M3 |
| LMArena Text: creative writing | 1370 | 1404 | MiniMax M3 |
| LMArena Vision | 1245 | 1251 | MiniMax M3 |
| LMArena Text: English | 1412 | 1443 | MiniMax M3 |
| LMArena Text: non-English | 1405 | 1420 | MiniMax M3 |
| LMArena Text: Chinese | 1446 | 1466 | MiniMax M3 |
| LMArena Text: German | 1408 | 1428 | MiniMax M3 |
| LMArena Text: Japanese | 1375 | 1380 | MiniMax M3 |
| LMArena Text: Korean | 1370 | 1373 | MiniMax M3 |
| LMArena Text: French | 1441 | 1444 | MiniMax M3 |
| LMArena Text: Spanish | 1405 | 1433 | MiniMax M3 |
| LMArena Text: Russian | 1417 | 1426 | MiniMax M3 |
| LMArena Text: Polish | 1415 | 1423 | MiniMax M3 |
| LMArena Text: multi-turn | 1429 | 1443 | MiniMax M3 |
| LMArena Text: longer queries | 1408 | 1446 | MiniMax M3 |
| LMArena Text: writing and language | 1393 | 1419 | MiniMax M3 |
| LMArena Text: entertainment and media | 1373 | 1403 | MiniMax M3 |
| LMArena Vision: English | 1244 | 1263 | MiniMax M3 |
| LMArena Vision: Chinese | 1286 | 1311 | MiniMax M3 |
| LMArena Vision: reading text | 1253 | 1259 | MiniMax M3 |
| LMArena Vision: humor | 1235 | 1235 | GPT-5.4 Mini |
| LMArena Vision: recognizing things | 1218 | 1197 | GPT-5.4 Mini |
| LMArena Vision: creative writing | 1222 | 1234 | MiniMax M3 |
| LMArena Document | Not measured | 1435 | Only MiniMax M3 measured |
| Neither model has a result on the other 11 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-5.4 Mini | MiniMax M3 | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
MiniMax M3 costs less
Repository review
50K input + 3K output tokens
MiniMax M3 costs less
Agent loop
200K input + 10K output tokens
MiniMax M3 costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | OpenAI GPT-5.4 Mini | MiniMax M3 |
|---|---|---|
| Maker | OpenAI | MiniMax |
| Weights | No public weights listed | Open weights |
| Released | March 17, 2026 | May 31, 2026 |
| Context window | 400K | 1.05M |
| Input $/Mtok | $0.75 | $0.30 |
| Output $/Mtok | $4.50 | $1.20 |
| Coding index | 28.2 | 57.8 |
| Independent measurements behind it | 1 | 2 |
| Cost per solved task | $0.70 | $0.56 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5.4 Mini better than MiniMax M3 for coding?
Pick MiniMax M3 for coding: it scores 57.8 on the Vibe Coding Index against 28.2 for GPT-5.4 Mini. Pick GPT-5.4 Mini when what matters is Terminal-Bench 4 (Vals AI) or Vibe Code Bench (Vals AI), where it leads.
Which is cheaper, GPT-5.4 Mini or MiniMax M3?
GPT-5.4 Mini costs $0.75 per million input tokens and $4.50 per million output tokens; MiniMax M3 costs $0.30 and $1.20. Per solved SWE-bench Verified task, GPT-5.4 Mini costs $0.70 and MiniMax M3 $0.56.
Which has the bigger context window, GPT-5.4 Mini or MiniMax M3?
MiniMax M3, with 1.05M tokens against 400K.
How many benchmarks is the GPT-5.4 Mini vs MiniMax M3 comparison based on?
78 benchmarks have a result for both models as of October 21, 2026. GPT-5.4 Mini's coding index rests on 1 independent benchmark and MiniMax M3's on 2. Where only one model has a result, the page says so instead of naming a winner.