Enter opens the highlighted result. Esc closes.

GPT-5.2 vs GPT-5 Mini

GPT-5.2 leads GPT-5 Mini on coding, 55.9 to 34.2 on the Vibe Coding Index, and GPT-5 Mini costs 7.0x less per output token ($2.00 vs $14.00 per million). Updated October 22, 2026; 96 benchmarks have a result for both.

Replace GPT-5.2 Replace GPT-5 Mini
GPT-5.2OpenAI
Coding
55.9
Overall
60.7
Output $/M
$14.00
92
GPT-5 MiniOpenAI
Coding
34.2
Overall
43.7
Output $/M
$2.00
3
vs boards won of 96 shared, 1 tied

GPT-5.2 or GPT-5 Mini: which to pick

Pick GPT-5.2 for coding: it scores 55.9 on the Vibe Coding Index against 34.2 for GPT-5 Mini.

Pick GPT-5 Mini for output tokens 7.0x cheaper ($2.00 vs $14.00 per million) and a lower cost per solved task ($0.08 against $1.91).

Model AGPT-5.2OpenAI55.9/1003 independent benchmarks$1.91 per solved taskModel BGPT-5 MiniOpenAI34.2/1002 independent benchmarks$0.08 per solved task
Shared results
96
GPT-5.2 only
18
GPT-5 Mini only
5
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

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 55.9 from 3 benchmarks, GPT-5 Mini 34.2 from 2 benchmarks.
Widest category
Reasoning, 17.0 points apart on average over 16 shared benchmarks.
Widest single gap
ARC-AGI-2: GPT-5.2 52.9% against GPT-5 Mini 4.4%, 48.5 points apart on a 0 to 100 scale.
Output price
GPT-5.2 $14.00 against GPT-5 Mini $2.00 per million output tokens, 7.0x apart.
Context window
Both take 400K tokens.

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-5.2BarsGPT-5 MiniTies
Coding101
Intelligence20
Agentic work30
Reasoning160
Math100
Knowledge132
Web design80
General3001 tie
All shared boards9231 tie
CodingLike-for-likeSix of these boards feed the Vibe Coding Index101
Coding benchmarks for GPT-5.2 and GPT-5 Mini
BenchmarkGPT-5.2GPT-5 MiniLeader
SWE-bench Verified72.8%59.8%GPT-5.2
SWE-bench Multilingual66.7%39.7%GPT-5.2
LMArena WebDev1417Not measuredOnly GPT-5.2 measured
LiveCodeBench (Vals AI)85.4%86.6%GPT-5 Mini
SWE-bench Verified (Vals AI)75.8%60.8%GPT-5.2
Vibe Code Bench (Vals AI)53.5%14.2%GPT-5.2
LiveCodeBench (Artificial Analysis run)89.4%83.8%GPT-5.2
SciCodeNot measured39.0%Only GPT-5 Mini measured
Terminal-Bench Hard (Artificial Analysis run)47.0%33.3%GPT-5.2
Terminal-Bench64.9%34.8%GPT-5.2
GSO-Bench27.5%Not measuredOnly GPT-5.2 measured
WeirdML72.2%52.7%GPT-5.2
ALE-Bench1293.5Not measuredOnly GPT-5.2 measured
LMArena Text: coding14421406GPT-5.2
LMArena Text: software and IT14341394GPT-5.2
LMArena WebDev: HTML1427Not measuredOnly GPT-5.2 measured
Neither model has a result on the other 23 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for GPT-5.2 and GPT-5 Mini
BenchmarkGPT-5.2GPT-5 MiniLeader
AA Intelligence Index30.420.6GPT-5.2
Epoch Capabilities Index153.7145.3GPT-5.2
Agentic workLike-for-likePublic boards, shown as published, kept out of the index30
Agentic work benchmarks for GPT-5.2 and GPT-5 Mini
BenchmarkGPT-5.2GPT-5 MiniLeader
Design Arena: agents building web apps1086Not measuredOnly GPT-5.2 measured
Design Arena: agents building full-stack apps1043Not measuredOnly GPT-5.2 measured
Design Arena: agents building mobile apps1093Not measuredOnly GPT-5.2 measured
Design Arena: agents building Android apps1059Not measuredOnly GPT-5.2 measured
Design Arena: agents building Godot games1142Not measuredOnly GPT-5.2 measured
τ²-Bench Telecom84.8%71.1%GPT-5.2
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
DeepResearch Bench41.1%Not measuredOnly GPT-5.2 measured
Vending-Bench 2$3,591-$31GPT-5.2
Berkeley Function Calling Leaderboard (BFCL)55.9%55.5%GPT-5.2
Neither model has a result on the other 29 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index160
Reasoning benchmarks for GPT-5.2 and GPT-5 Mini
BenchmarkGPT-5.2GPT-5 MiniLeader
GPQA Diamond91.7%80.3%GPT-5.2
MMMU Pro (Vals AI)86.7%78.9%GPT-5.2
GPQA Diamond (Artificial Analysis run)90.3%82.8%GPT-5.2
Humanity's Last Exam (Artificial Analysis run)37.7%21.5%GPT-5.2
GPQA Diamond (Epoch AI run)91.4%75.0%GPT-5.2
ARC-AGI86.2%54.3%GPT-5.2
ARC-AGI-252.9%4.4%GPT-5.2
SimpleBench45.8%Not measuredOnly GPT-5.2 measured
Chess puzzles48.0%24.0%GPT-5.2
Mystery game puzzles23.0%10.0%GPT-5.2
EBR-bench23.0%Not measuredOnly GPT-5.2 measured
VPCT84.0%40.2%GPT-5.2
LMArena Text: hard prompts14281380GPT-5.2
LMArena Text: expert14451379GPT-5.2
LMArena Text: hard prompts in English14311383GPT-5.2
LMArena Vision: diagrams12651202GPT-5.2
LMArena Vision: homework12761210GPT-5.2
Kagi LLM Benchmark73.3%70.3%GPT-5.2
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index100
Math benchmarks for GPT-5.2 and GPT-5 Mini
BenchmarkGPT-5.2GPT-5 MiniLeader
AIME96.9%91.5%GPT-5.2
MATH 500 (Vals AI)Not measured94.8%Only GPT-5 Mini measured
MGSM (Vals AI)94.0%92.6%GPT-5.2
AA Math Index99.090.7GPT-5.2
AIME 2025 (Artificial Analysis run)99.0%90.7%GPT-5.2
OTIS Mock AIME96.1%86.7%GPT-5.2
FrontierMath, tiers 1 to 367.4%46.7%GPT-5.2
FrontierMath, tier 431.7%12.2%GPT-5.2
ProofBench15.0%9.0%GPT-5.2
MATH level 5Not measured97.8%Only GPT-5 Mini measured
LMArena Text: math14401378GPT-5.2
LMArena Text: mathematical work14421380GPT-5.2
Neither model has a result on the other 1 board here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index132
Knowledge benchmarks for GPT-5.2 and GPT-5 Mini
BenchmarkGPT-5.2GPT-5 MiniLeader
MMLU Pro86.2%82.2%GPT-5.2
LegalBench (Vals AI)82.8%81.8%GPT-5.2
CaseLaw (Vals AI)66.0%68.5%GPT-5 Mini
CorpFin (Vals AI)65.9%60.2%GPT-5.2
TaxEval (Vals AI)75.8%75.2%GPT-5.2
MortgageTax (Vals AI)67.1%66.9%GPT-5.2
MedQA (Vals AI)94.1%96.1%GPT-5 Mini
MedCode (Vals AI)49.8%43.0%GPT-5.2
MMLU Pro (Artificial Analysis run)87.4%83.7%GPT-5.2
SimpleQA Verified37.1%21.6%GPT-5.2
LMArena Text: science14251371GPT-5.2
LMArena Text: medicine and healthcare14251380GPT-5.2
LMArena Text: legal and government14271375GPT-5.2
LMArena Text: business and finance14141358GPT-5.2
FACTS Benchmark Suite48.5%38.6%GPT-5.2
Neither model has a result on the other 3 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index80
Web design benchmarks for GPT-5.2 and GPT-5 Mini
BenchmarkGPT-5.2GPT-5 MiniLeader
Design Arena: websites11951125GPT-5.2
Design Arena: UI components11901103GPT-5.2
Design Arena: game dev11951129GPT-5.2
Design Arena: data viz12021131GPT-5.2
Design Arena: 3D10791036GPT-5.2
Design Arena: all code categories11701114GPT-5.2
Design Arena: SVG11391091GPT-5.2
Design Arena: ASCII art11521125GPT-5.2
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index300
General benchmarks for GPT-5.2 and GPT-5 Mini
BenchmarkGPT-5.2GPT-5 MiniLeader
SAGE (Vals AI)49.3%43.0%GPT-5.2
MedScribe (Vals AI)84.4%80.6%GPT-5.2
IFBench75.4%75.4%Tie
Long context reasoning (AA-LCR)82.7%72.3%GPT-5.2
DTBench90.9%80.5%GPT-5.2
LMCA43.9%34.2%GPT-5.2
Fiction.LiveBenchNot measured69.4%Only GPT-5 Mini measured
Creative writing (Lech Mazur)Not measured83.1%Only GPT-5 Mini measured
CL-bench18.2%Not measuredOnly GPT-5.2 measured
Furniture assembly38.3%Not measuredOnly GPT-5.2 measured
LMArena Text14161373GPT-5.2
LMArena Text: instruction following14091357GPT-5.2
LMArena Text: creative writing13761325GPT-5.2
LMArena Vision12461202GPT-5.2
LMArena Text: English14221379GPT-5.2
LMArena Text: non-English14051363GPT-5.2
LMArena Text: Chinese14471386GPT-5.2
LMArena Text: German14441366GPT-5.2
LMArena Text: Japanese14201341GPT-5.2
LMArena Text: Korean13741308GPT-5.2
LMArena Text: French14541386GPT-5.2
LMArena Text: Spanish14131354GPT-5.2
LMArena Text: Russian14151362GPT-5.2
LMArena Text: Polish14261378GPT-5.2
LMArena Text: multi-turn14191363GPT-5.2
LMArena Text: longer queries14131355GPT-5.2
LMArena Text: writing and language13921344GPT-5.2
LMArena Text: entertainment and media13781340GPT-5.2
LMArena Vision: English12431216GPT-5.2
LMArena Vision: Chinese13051214GPT-5.2
LMArena Vision: reading text12571199GPT-5.2
LMArena Vision: humor12271205GPT-5.2
LMArena Vision: captioning12671217GPT-5.2
LMArena Vision: recognizing things12321226GPT-5.2
LMArena Vision: creative writing12251181GPT-5.2
LMArena Document1405Not measuredOnly GPT-5.2 measured
Neither model has a result on the other 10 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 GPT-5 Mini
BenchmarkGPT-5.2GPT-5 MiniLeader
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
GPT-5 Mini$0.0013Fits in one request

GPT-5 Mini costs less

Repository review

50K input + 3K output tokens

GPT-5.2$0.130Fits in one request
GPT-5 Mini$0.018Fits in one request

GPT-5 Mini costs less

Agent loop

200K input + 10K output tokens

GPT-5.2$0.490Fits in one request
GPT-5 Mini$0.070Fits in one request

GPT-5 Mini 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 GPT-5 Mini
SpecificationOpenAI GPT-5.2OpenAI GPT-5 Mini
MakerOpenAIOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedDecember 10, 2025August 7, 2025
Context window400K400K
Input $/Mtok$1.75$0.25
Output $/Mtok$14.00$2.00
Coding index55.934.2
Independent measurements behind it32
Cost per solved task$1.91$0.08
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-5.2 better than GPT-5 Mini for coding?

Pick GPT-5.2 for coding: it scores 55.9 on the Vibe Coding Index against 34.2 for GPT-5 Mini. Pick GPT-5 Mini for output tokens 7.0x cheaper ($2.00 vs $14.00 per million) and a lower cost per solved task ($0.08 against $1.91).

Which is cheaper, GPT-5.2 or GPT-5 Mini?

GPT-5.2 costs $1.75 per million input tokens and $14.00 per million output tokens; GPT-5 Mini costs $0.25 and $2.00. Per solved SWE-bench Verified task, GPT-5.2 costs $1.91 and GPT-5 Mini $0.08.

Which has the bigger context window, GPT-5.2 or GPT-5 Mini?

Neither: both take 400K tokens.

How many benchmarks is the GPT-5.2 vs GPT-5 Mini comparison based on?

96 benchmarks have a result for both models as of October 22, 2026. GPT-5.2's coding index rests on 3 independent benchmarks and GPT-5 Mini'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-gpt-5-mini" title="GPT-5.2 vs GPT-5 Mini, 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-gpt-5-mini">LLMBenchmark.ai, GPT-5.2 vs GPT-5 Mini</a></p>