LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

GPT-5.2 vs MiniMax M3

Updated October 21, 2026. 83 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GPT-5.2 Replace MiniMax M3
GPT-5.2OpenAI
Coding
58.9
Overall
30.4
Output $/M
$14.00
27
MiniMax M3MiniMax
Coding
57.8
Overall
29.2
Output $/M
$1.20
56
vs boards won of 83 shared

Decision reading

GPT-5.2 and MiniMax M3 are level for coding, 58.9 against 57.8 on the Vibe Coding Index, so price and context decide.

Pick MiniMax M3 when output tokens cost 11.7x less, or when a solved task costs $0.56 against $1.91. Pick GPT-5.2 when what matters is LiveCodeBench (Vals AI) or SWE-bench Verified (Vals AI), where it leads.

Model AGPT-5.2OpenAI58.9/1002 independent benchmarks$1.91 per solved taskModel BMiniMax M3MiniMax57.8/1002 independent benchmarks$0.56 per solved task
Shared results
83
GPT-5.2 only
28
MiniMax M3 only
33
Like-for-like categories
8 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.

At a glance

Coding index
GPT-5.2 58.9 from 2 benchmarks, MiniMax M3 57.8 from 2 benchmarks.
Widest category
Agentic work, 11.4 points apart on average over 6 shared benchmarks.
Widest single gap
Chess puzzles: GPT-5.2 48.0% against MiniMax M3 13.0%, 35.0 points apart on a 0 to 100 scale.
Output price
GPT-5.2 $14.00 against MiniMax M3 $1.20 per million output tokens, 11.7x apart.
Context window
GPT-5.2 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.

Boards won, where both have a result
CategoryGPT-5.2BarsMiniMax M3Ties
Coding54
Intelligence20
Agentic work15
Reasoning67
Math22
Knowledge37
Web design08
General823
All shared boards2756
CodingLike-for-likeFour of these boards feed the Vibe Coding Index54
Coding benchmarks for GPT-5.2 and MiniMax M3
BenchmarkGPT-5.2MiniMax M3Leader
SWE-bench Verified72.8%Not measuredOnly GPT-5.2 measured
SWE-bench Multilingual66.7%Not measuredOnly GPT-5.2 measured
LiveBench codingNot measured70.9%Only MiniMax M3 measured
LMArena WebDev14171482MiniMax M3
AA Coding IndexNot measured58.6Only MiniMax M3 measured
LiveCodeBench (Vals AI)85.4%82.2%GPT-5.2
SWE-bench Verified (Vals AI)75.8%75.0%GPT-5.2
Terminal-Bench 4 (Vals AI)Not measured1.0%Only MiniMax M3 measured
Vibe Code Bench (Vals AI)53.5%47.6%GPT-5.2
Code migration (Vals AI)Not measured19.9%Only MiniMax M3 measured
Terminal-Bench 2.1 (Vals AI)Not measured53.6%Only MiniMax M3 measured
SWE-bench Multimodal (Vals AI)Not measured33.3%Only MiniMax M3 measured
Vibe Code Bench 1-100 (Vals AI)Not measured9.2%Only MiniMax M3 measured
SkillsBench (Vals AI)Not measured51.5%Only MiniMax M3 measured
CyberBench (Vals AI)Not measured65.0%Only MiniMax M3 measured
LiveCodeBench (Artificial Analysis run)89.4%Not measuredOnly GPT-5.2 measured
SciCodeNot measured47.1%Only MiniMax M3 measured
Terminal-Bench Hard (Artificial Analysis run)47.0%42.4%GPT-5.2
Terminal-Bench64.9%Not measuredOnly GPT-5.2 measured
FrontierCodeNot measured14.7%Only MiniMax M3 measured
GSO-Bench27.5%Not measuredOnly GPT-5.2 measured
WeirdML72.2%Not measuredOnly GPT-5.2 measured
ALE-Bench1293.5640.0GPT-5.2
LMArena Text: coding14421470MiniMax M3
LMArena Text: software and IT14341466MiniMax M3
LMArena WebDev: ReactNot measured1478Only MiniMax M3 measured
LMArena WebDev: HTML14271455MiniMax M3
LMArena WebDev: from an imageNot measured1470Only MiniMax M3 measured
Neither model has a result on the other 10 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for GPT-5.2 and MiniMax M3
BenchmarkGPT-5.2MiniMax M3Leader
AA Intelligence Index30.429.2GPT-5.2
Epoch Capabilities Index153.7146.9GPT-5.2
Agentic workLike-for-likePublic boards, shown as published, kept out of the index15
Agentic work benchmarks for GPT-5.2 and MiniMax M3
BenchmarkGPT-5.2MiniMax M3Leader
AA Agentic IndexNot measured29.5Only MiniMax M3 measured
Design Arena: agents building web apps10861200MiniMax M3
Design Arena: agents building full-stack apps10431197MiniMax M3
Design Arena: agents building mobile apps10941169MiniMax M3
Design Arena: agents building Android apps10641164MiniMax M3
Design Arena: agents building Godot games1142Not measuredOnly GPT-5.2 measured
Design Arena: agents building gamesNot measured1158Only MiniMax M3 measured
Design Arena: HTML slidesNot measured1180Only MiniMax M3 measured
Design Arena: PowerPoint slidesNot measured1205Only MiniMax M3 measured
Design Arena: spreadsheetsNot measured1021Only MiniMax M3 measured
LiveBench researchNot measured68.0%Only MiniMax M3 measured
LiveBench browserNot measured8.0%Only MiniMax M3 measured
LiveBench long agenticNot measured12.4%Only MiniMax M3 measured
Finance Agent (Vals AI)Not measured48.3%Only MiniMax M3 measured
Tax Agent Bench (Vals AI)Not measured20.6%Only MiniMax M3 measured
Legal Research Bench (Vals AI)Not measured29.8%Only MiniMax M3 measured
Excel Modeling Benchmark (Vals AI)Not measured47.8%Only MiniMax M3 measured
τ²-Bench Telecom84.8%88.9%MiniMax M3
APEX-AgentsNot measured37.7%Only MiniMax M3 measured
METR time horizons75.3%Not measuredOnly GPT-5.2 measured
Remote Labor Index2.5%Not measuredOnly GPT-5.2 measured
GDPval49.7%Not measuredOnly GPT-5.2 measured
OSWorld 2.0Not measured4.6%Only MiniMax M3 measured
DeepResearch Bench41.1%Not measuredOnly GPT-5.2 measured
Vending-Bench 2$3,591$2,158GPT-5.2
Neither model has a result on the other 14 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index67
Reasoning benchmarks for GPT-5.2 and MiniMax M3
BenchmarkGPT-5.2MiniMax M3Leader
GPQA Diamond91.7%92.7%MiniMax M3
LiveBench reasoningNot measured74.5%Only MiniMax M3 measured
MMMU Pro (Vals AI)86.7%81.2%GPT-5.2
GPQA Diamond (Artificial Analysis run)90.3%92.9%MiniMax M3
Humanity's Last Exam (Artificial Analysis run)37.7%39.0%MiniMax M3
GPQA Diamond (Epoch AI run)91.4%90.9%GPT-5.2
ARC-AGI86.2%Not measuredOnly GPT-5.2 measured
ARC-AGI-252.9%Not measuredOnly GPT-5.2 measured
SimpleBench45.8%45.8%GPT-5.2
Chess puzzles48.0%13.0%GPT-5.2
Mystery game puzzles23.0%8.0%GPT-5.2
EBR-bench23.0%Not measuredOnly GPT-5.2 measured
VPCT84.0%Not measuredOnly GPT-5.2 measured
LMArena Text: hard prompts14281447MiniMax M3
LMArena Text: expert14451461MiniMax M3
LMArena Text: hard prompts in English14311455MiniMax M3
LMArena Vision: diagrams12651266MiniMax M3
LMArena Vision: homework12761268GPT-5.2
Neither model has a result on the other 3 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index22
Math benchmarks for GPT-5.2 and MiniMax M3
BenchmarkGPT-5.2MiniMax M3Leader
AIME96.9%Not measuredOnly GPT-5.2 measured
MGSM (Vals AI)94.0%Not measuredOnly GPT-5.2 measured
AA Math Index99.0Not measuredOnly GPT-5.2 measured
AIME 2025 (Artificial Analysis run)99.0%Not measuredOnly GPT-5.2 measured
OTIS Mock AIME96.1%71.1%GPT-5.2
FrontierMath, tiers 1 to 367.4%Not measuredOnly GPT-5.2 measured
FrontierMath, tier 431.7%Not measuredOnly GPT-5.2 measured
ProofBench15.0%18.0%MiniMax M3
LMArena Text: math14401429GPT-5.2
LMArena Text: mathematical work14421448MiniMax M3
Neither model has a result on the other 3 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index37
Knowledge benchmarks for GPT-5.2 and MiniMax M3
BenchmarkGPT-5.2MiniMax M3Leader
MMLU Pro86.2%84.2%GPT-5.2
LegalBench (Vals AI)82.8%85.4%MiniMax M3
CaseLaw (Vals AI)66.0%Not measuredOnly GPT-5.2 measured
CorpFin (Vals AI)65.9%68.1%MiniMax M3
TaxEval (Vals AI)75.8%72.7%GPT-5.2
MortgageTax (Vals AI)67.1%68.4%MiniMax M3
MedQA (Vals AI)94.1%Not measuredOnly GPT-5.2 measured
MedCode (Vals AI)49.8%46.3%GPT-5.2
MMLU Pro (Artificial Analysis run)87.4%Not measuredOnly GPT-5.2 measured
SimpleQA Verified37.1%Not measuredOnly GPT-5.2 measured
LMArena Text: science14251452MiniMax M3
LMArena Text: medicine and healthcare14251441MiniMax M3
LMArena Text: legal and government14271437MiniMax M3
LMArena Text: business and finance14141434MiniMax M3
Neither model has a result on the other 3 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index08
Web design benchmarks for GPT-5.2 and MiniMax M3
BenchmarkGPT-5.2MiniMax M3Leader
Design Arena: websites11961257MiniMax M3
Design Arena: UI components11901246MiniMax M3
Design Arena: game dev11961210MiniMax M3
Design Arena: data viz12021235MiniMax M3
Design Arena: 3D10791209MiniMax M3
Design Arena: all code categories11711247MiniMax M3
Design Arena: SVG11371161MiniMax M3
Design Arena: ASCII art11531161MiniMax M3
LiveBench designNot measured66.2%Only MiniMax M3 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index823
General benchmarks for GPT-5.2 and MiniMax M3
BenchmarkGPT-5.2MiniMax M3Leader
LiveBench data analysisNot measured76.2%Only MiniMax M3 measured
Public Benefits Bench (Vals AI)Not measured64.1%Only MiniMax M3 measured
SAGE (Vals AI)49.3%50.6%MiniMax M3
MedScribe (Vals AI)84.4%87.3%MiniMax M3
Vals Multimodal IndexNot measured60.0%Only MiniMax M3 measured
IFBench75.4%82.9%MiniMax M3
Long context reasoning (AA-LCR)82.7%83.0%MiniMax M3
DTBench90.9%78.9%GPT-5.2
LMCA43.9%33.7%GPT-5.2
CL-bench18.2%Not measuredOnly GPT-5.2 measured
Surface Evolver BenchNot measured55.0%Only MiniMax M3 measured
Furniture assembly38.3%Not measuredOnly GPT-5.2 measured
LMArena Text14161433MiniMax M3
LMArena Text: instruction following14091433MiniMax M3
LMArena Text: creative writing13761404MiniMax M3
LMArena Vision12461251MiniMax M3
LMArena Text: English14221443MiniMax M3
LMArena Text: non-English14051420MiniMax M3
LMArena Text: Chinese14471466MiniMax M3
LMArena Text: German14441428GPT-5.2
LMArena Text: Japanese14201380GPT-5.2
LMArena Text: Korean13741373GPT-5.2
LMArena Text: French14541444GPT-5.2
LMArena Text: Spanish14131433MiniMax M3
LMArena Text: Russian14151426MiniMax M3
LMArena Text: Polish14261423GPT-5.2
LMArena Text: multi-turn14191443MiniMax M3
LMArena Text: longer queries14131446MiniMax M3
LMArena Text: writing and language13921419MiniMax M3
LMArena Text: entertainment and media13781403MiniMax M3
LMArena Vision: English12431263MiniMax M3
LMArena Vision: Chinese13051311MiniMax M3
LMArena Vision: reading text12571259MiniMax M3
LMArena Vision: humor12271235MiniMax M3
LMArena Vision: captioning1267Not measuredOnly GPT-5.2 measured
LMArena Vision: recognizing things12321197GPT-5.2
LMArena Vision: creative writing12251234MiniMax M3
LMArena Document14051435MiniMax M3
Neither model has a result on the other 8 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-5.2 and MiniMax M3
BenchmarkGPT-5.2MiniMax M3Leader
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

GPT-5.2$0.0088Fits in one request
MiniMax M3$0.0009Fits in one request

MiniMax M3 costs less

Repository review

50K input + 3K output tokens

GPT-5.2$0.130Fits in one request
MiniMax M3$0.019Fits in one request

MiniMax M3 costs less

Agent loop

200K input + 10K output tokens

GPT-5.2$0.490Fits in one request
MiniMax M3$0.072Fits in one request

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.

Specifications of GPT-5.2 and MiniMax M3
SpecificationOpenAI GPT-5.2MiniMax M3
MakerOpenAIMiniMax
WeightsNo public weights listedOpen weights
ReleasedDecember 10, 2025May 31, 2026
Context window400K1.05M
Input $/Mtok$1.75$0.30
Output $/Mtok$14.00$1.20
Coding index58.957.8
Independent measurements behind it22
Cost per solved task$1.91$0.56
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-5.2 better than MiniMax M3 for coding?

GPT-5.2 and MiniMax M3 are level for coding, 58.9 against 57.8 on the Vibe Coding Index, so price and context decide. Pick MiniMax M3 when output tokens cost 11.7x less, or when a solved task costs $0.56 against $1.91. Pick GPT-5.2 when what matters is LiveCodeBench (Vals AI) or SWE-bench Verified (Vals AI), where it leads.

Which is cheaper, GPT-5.2 or MiniMax M3?

GPT-5.2 costs $1.75 per million input tokens and $14.00 per million output tokens; MiniMax M3 costs $0.30 and $1.20. Per solved SWE-bench Verified task, GPT-5.2 costs $1.91 and MiniMax M3 $0.56.

Which has the bigger context window, GPT-5.2 or MiniMax M3?

MiniMax M3, with 1.05M tokens against 400K.

How many benchmarks is the GPT-5.2 vs MiniMax M3 comparison based on?

83 benchmarks have a result for both models as of October 21, 2026. GPT-5.2's coding index rests on 2 independent benchmarks and MiniMax M3's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/gpt-5-2-vs-minimax-minimax-m3" title="GPT-5.2 vs MiniMax M3, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/gpt-5-2-vs-minimax-minimax-m3">LLMBenchmark.ai, GPT-5.2 vs MiniMax M3</a></p>