Enter opens the highlighted result. Esc closes.

GPT-4.1 vs GPT-5.2

GPT-5.2 leads GPT-4.1 on the Artificial Analysis Intelligence Index, 30.4 to 12.7, and GPT-4.1 costs 1.8x less per output token ($8.00 vs $14.00 per million). Updated October 21, 2026; 84 benchmarks have a result for both.

Replace GPT-4.1 Replace GPT-5.2
GPT-4.1OpenAI
Coding
No data
Overall
36.5
Output $/M
$8.00
2
GPT-5.2OpenAI
Coding
55.9
Overall
60.7
Output $/M
$14.00
82
vs boards won of 84 shared

GPT-4.1 or GPT-5.2: which to pick

Only GPT-5.2 has a coding score, 55.9 on the Vibe Coding Index; GPT-4.1 has none, so the two cannot be ranked on coding.

Pick GPT-4.1 for output tokens 1.8x cheaper ($8.00 vs $14.00 per million) and a lower cost per solved task ($0.37 against $1.91).

Model AGPT-4.1OpenAINot benchmarkedNo coding benchmark$0.37 per solved taskModel BGPT-5.2OpenAI55.9/1003 independent benchmarks$1.91 per solved task
Shared results
84
GPT-4.1 only
8
GPT-5.2 only
30
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-4.1 not benchmarked, GPT-5.2 55.9 from 3 benchmarks.
Widest category
Math, 42.1 points apart on average over 8 shared benchmarks.
Widest single gap
ARC-AGI: GPT-4.1 5.5% against GPT-5.2 86.2%, 80.7 points apart on a 0 to 100 scale.
Output price
GPT-4.1 $8.00 against GPT-5.2 $14.00 per million output tokens, 1.8x apart.
Context window
GPT-4.1 takes 1.05M tokens, GPT-5.2 400K.

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-4.1BarsGPT-5.2Ties
Coding08
Intelligence02
Agentic work02
Reasoning015
Math08
Knowledge212
Web design06
General029
All shared boards282
CodingLike-for-likeSix of these boards feed the Vibe Coding Index08
Coding benchmarks for GPT-4.1 and GPT-5.2
BenchmarkGPT-4.1GPT-5.2Leader
Aider polyglot52.4%Not measuredOnly GPT-4.1 measured
SWE-bench Verified39.6%72.8%GPT-5.2
SWE-bench Multimodal31.1%Not measuredOnly GPT-4.1 measured
SWE-bench MultilingualNot measured66.7%Only GPT-5.2 measured
LMArena WebDevNot measured1417Only GPT-5.2 measured
LiveCodeBench (Vals AI)54.7%85.4%GPT-5.2
SWE-bench Verified (Vals AI)Not measured75.8%Only GPT-5.2 measured
Vibe Code Bench (Vals AI)Not measured53.5%Only GPT-5.2 measured
LiveCodeBench (Artificial Analysis run)45.7%89.4%GPT-5.2
Terminal-Bench Hard (Artificial Analysis run)13.6%47.0%GPT-5.2
Terminal-BenchNot measured64.9%Only GPT-5.2 measured
GSO-BenchNot measured27.5%Only GPT-5.2 measured
WeirdML39.0%72.2%GPT-5.2
ALE-Bench558.11293.5GPT-5.2
LMArena Text: coding13901442GPT-5.2
LMArena Text: software and IT13911434GPT-5.2
LMArena WebDev: HTMLNot measured1427Only GPT-5.2 measured
SciCode4.6%Not measuredOnly GPT-4.1 measured
Neither model has a result on the other 21 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for GPT-4.1 and GPT-5.2
BenchmarkGPT-4.1GPT-5.2Leader
AA Intelligence Index12.730.4GPT-5.2
Epoch Capabilities Index136.7153.7GPT-5.2
Agentic workLike-for-likePublic boards, shown as published, kept out of the index02
Agentic work benchmarks for GPT-4.1 and GPT-5.2
BenchmarkGPT-4.1GPT-5.2Leader
Design Arena: agents building web appsNot measured1086Only GPT-5.2 measured
Design Arena: agents building full-stack appsNot measured1043Only GPT-5.2 measured
Design Arena: agents building mobile appsNot measured1093Only GPT-5.2 measured
Design Arena: agents building Android appsNot measured1058Only GPT-5.2 measured
Design Arena: agents building Godot gamesNot measured1142Only GPT-5.2 measured
τ²-Bench Telecom47.1%84.8%GPT-5.2
METR time horizonsNot measured75.3%Only GPT-5.2 measured
Remote Labor IndexNot measured2.5%Only GPT-5.2 measured
GDPvalNot measured49.7%Only GPT-5.2 measured
DeepResearch BenchNot measured41.1%Only GPT-5.2 measured
Vending-Bench 2Not measured$3,591Only GPT-5.2 measured
Berkeley Function Calling Leaderboard (BFCL)54.0%55.9%GPT-5.2
BrowseComp-Plus36.4%Not measuredOnly GPT-4.1 measured
Neither model has a result on the other 28 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index015
Reasoning benchmarks for GPT-4.1 and GPT-5.2
BenchmarkGPT-4.1GPT-5.2Leader
GPQA Diamond65.4%91.7%GPT-5.2
MMMU Pro (Vals AI)72.4%86.7%GPT-5.2
GPQA Diamond (Artificial Analysis run)66.6%90.3%GPT-5.2
Humanity's Last Exam (Artificial Analysis run)4.2%37.7%GPT-5.2
GPQA Diamond (Epoch AI run)66.9%91.4%GPT-5.2
ARC-AGI5.5%86.2%GPT-5.2
ARC-AGI-20.4%52.9%GPT-5.2
SimpleBench27.0%45.8%GPT-5.2
Chess puzzles3.0%48.0%GPT-5.2
Mystery game puzzlesNot measured23.0%Only GPT-5.2 measured
EBR-benchNot measured23.0%Only GPT-5.2 measured
VPCTNot measured84.0%Only GPT-5.2 measured
LMArena Text: hard prompts13841428GPT-5.2
LMArena Text: expert13641445GPT-5.2
LMArena Text: hard prompts in English13931431GPT-5.2
LMArena Vision: diagrams12151265GPT-5.2
LMArena Vision: homework12481276GPT-5.2
Kagi LLM Benchmark52.3%73.3%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 index08
Math benchmarks for GPT-4.1 and GPT-5.2
BenchmarkGPT-4.1GPT-5.2Leader
AIME39.6%96.9%GPT-5.2
MATH 500 (Vals AI)87.2%Not measuredOnly GPT-4.1 measured
MGSM (Vals AI)87.7%94.0%GPT-5.2
AA Math Index34.799.0GPT-5.2
AIME 2025 (Artificial Analysis run)34.7%99.0%GPT-5.2
OTIS Mock AIME38.3%96.1%GPT-5.2
FrontierMath, tiers 1 to 36.0%67.4%GPT-5.2
FrontierMath, tier 4Not measured31.7%Only GPT-5.2 measured
ProofBenchNot measured15.0%Only GPT-5.2 measured
MATH level 583.0%Not measuredOnly GPT-4.1 measured
LMArena Text: math13691440GPT-5.2
LMArena Text: mathematical work13601442GPT-5.2
Neither model has a result on the other 1 board here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index212
Knowledge benchmarks for GPT-4.1 and GPT-5.2
BenchmarkGPT-4.1GPT-5.2Leader
MMLU Pro80.5%86.2%GPT-5.2
LegalBench (Vals AI)83.1%82.8%GPT-4.1
CaseLaw (Vals AI)69.9%66.0%GPT-4.1
CorpFin (Vals AI)63.0%65.9%GPT-5.2
TaxEval (Vals AI)75.1%75.8%GPT-5.2
MortgageTax (Vals AI)65.9%67.1%GPT-5.2
MedQA (Vals AI)91.2%94.1%GPT-5.2
MedCode (Vals AI)Not measured49.8%Only GPT-5.2 measured
MMLU Pro (Artificial Analysis run)80.6%87.4%GPT-5.2
SimpleQA Verified31.1%37.1%GPT-5.2
GeoBench72.0%Not measuredOnly GPT-4.1 measured
LMArena Text: science13821425GPT-5.2
LMArena Text: medicine and healthcare13791425GPT-5.2
LMArena Text: legal and government13991427GPT-5.2
LMArena Text: business and finance13691414GPT-5.2
FACTS Benchmark Suite45.8%48.5%GPT-5.2
Neither model has a result on the other 2 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index06
Web design benchmarks for GPT-4.1 and GPT-5.2
BenchmarkGPT-4.1GPT-5.2Leader
Design Arena: websites10391195GPT-5.2
Design Arena: UI components10001190GPT-5.2
Design Arena: game dev10801195GPT-5.2
Design Arena: data viz11041202GPT-5.2
Design Arena: 3D8501079GPT-5.2
Design Arena: all code categories10261170GPT-5.2
Design Arena: SVGNot measured1139Only GPT-5.2 measured
Design Arena: ASCII artNot measured1152Only GPT-5.2 measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index029
General benchmarks for GPT-4.1 and GPT-5.2
BenchmarkGPT-4.1GPT-5.2Leader
SAGE (Vals AI)Not measured49.3%Only GPT-5.2 measured
MedScribe (Vals AI)Not measured84.4%Only GPT-5.2 measured
IFBench43.0%75.4%GPT-5.2
Long context reasoning (AA-LCR)68.3%82.7%GPT-5.2
DTBench68.3%90.9%GPT-5.2
LMCA25.6%43.9%GPT-5.2
Fiction.LiveBench63.9%Not measuredOnly GPT-4.1 measured
CL-benchNot measured18.2%Only GPT-5.2 measured
Furniture assemblyNot measured38.3%Only GPT-5.2 measured
LMArena Text13831416GPT-5.2
LMArena Text: instruction following13671409GPT-5.2
LMArena Text: creative writing13631376GPT-5.2
LMArena Vision12101246GPT-5.2
LMArena Text: English13921422GPT-5.2
LMArena Text: non-English13701405GPT-5.2
LMArena Text: Chinese13821447GPT-5.2
LMArena Text: German13811444GPT-5.2
LMArena Text: Japanese13191420GPT-5.2
LMArena Text: Korean13381374GPT-5.2
LMArena Text: French13821454GPT-5.2
LMArena Text: Spanish13761413GPT-5.2
LMArena Text: Russian13771415GPT-5.2
LMArena Text: Polish13891426GPT-5.2
LMArena Text: multi-turn13971419GPT-5.2
LMArena Text: longer queries13851413GPT-5.2
LMArena Text: writing and language13711392GPT-5.2
LMArena Text: entertainment and media13651378GPT-5.2
LMArena Vision: English12091243GPT-5.2
LMArena Vision: Chinese12261305GPT-5.2
LMArena Vision: reading text12181257GPT-5.2
LMArena Vision: humor11961227GPT-5.2
LMArena Vision: captioning12071267GPT-5.2
LMArena Vision: recognizing things12081232GPT-5.2
LMArena Vision: creative writing12101225GPT-5.2
LMArena DocumentNot measured1405Only GPT-5.2 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
Image generation benchmarks for GPT-4.1 and GPT-5.2
BenchmarkGPT-4.1GPT-5.2Leader
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-4.1$0.0060Fits in one request
GPT-5.2$0.0088Fits in one request

GPT-4.1 costs less

Repository review

50K input + 3K output tokens

GPT-4.1$0.124Fits in one request
GPT-5.2$0.130Fits in one request

GPT-4.1 costs less

Agent loop

200K input + 10K output tokens

GPT-4.1$0.480Fits in one request
GPT-5.2$0.490Fits in one request

GPT-4.1 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-4.1 and GPT-5.2
SpecificationOpenAI GPT-4.1OpenAI GPT-5.2
MakerOpenAIOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedApril 14, 2025December 10, 2025
Context window1.05M400K
Input $/Mtok$2.00$1.75
Output $/Mtok$8.00$14.00
Coding indexNot benchmarked55.9
Independent measurements behind it03
Cost per solved task$0.37$1.91
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-4.1 better than GPT-5.2 for coding?

Only GPT-5.2 has a coding score, 55.9 on the Vibe Coding Index; GPT-4.1 has none, so the two cannot be ranked on coding. Pick GPT-4.1 for output tokens 1.8x cheaper ($8.00 vs $14.00 per million) and a lower cost per solved task ($0.37 against $1.91).

Which is cheaper, GPT-4.1 or GPT-5.2?

GPT-4.1 costs $2.00 per million input tokens and $8.00 per million output tokens; GPT-5.2 costs $1.75 and $14.00. Per solved SWE-bench Verified task, GPT-4.1 costs $0.37 and GPT-5.2 $1.91.

Which has the bigger context window, GPT-4.1 or GPT-5.2?

GPT-4.1, with 1.05M tokens against 400K.

How many benchmarks is the GPT-4.1 vs GPT-5.2 comparison based on?

84 benchmarks have a result for both models as of October 21, 2026. GPT-4.1's coding index rests on 0 independent benchmarks and GPT-5.2's on 3. Where only one model has a result, the page says so instead of naming a winner.

Compare next