Enter opens the highlighted result. Esc closes.

Gemma 3 27B vs GPT-4.1

Updated October 21, 2026. 58 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Gemma 3 27B Replace GPT-4.1
Gemma 3 27BGoogle
Coding
4.9
Overall
4.9
Output $/M
$0.35
3
GPT-4.1OpenAI
Coding
46.5
Overall
12.7
Output $/M
$8.00
55
vs boards won of 58 shared

Decision reading

Pick GPT-4.1 for coding: it scores 46.5 on the Vibe Coding Index against 4.9 for Gemma 3 27B.

Pick Gemma 3 27B when output tokens cost 22.9x less, or when what matters is Humanity's Last Exam (Artificial Analysis run) or LMArena Text: science, where it leads.

Model AGemma 3 27BGoogle4.9/1001 independent benchmarkNo cost per solved taskModel BGPT-4.1OpenAI46.5/1002 independent benchmarks$0.37 per solved task
Shared results
58
Gemma 3 27B only
4
GPT-4.1 only
29
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.

At a glance

Coding index
Gemma 3 27B 4.9 from 1 benchmark, GPT-4.1 46.5 from 2 benchmarks.
Widest category
Agentic work, 36.5 points apart on average over 1 shared benchmark.
Widest single gap
Long context reasoning (AA-LCR): Gemma 3 27B 7.3% against GPT-4.1 68.3%, 61.0 points apart on a 0 to 100 scale.
Output price
Gemma 3 27B $0.35 against GPT-4.1 $8.00 per million output tokens, 22.9x apart.
Context window
Gemma 3 27B takes 131K tokens, GPT-4.1 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
CategoryGemma 3 27BBarsGPT-4.1Ties
Coding05
Intelligence02
Agentic work01
Reasoning17
Math06
Knowledge24
General030
All shared boards355
CodingLike-for-likeFour of these boards feed the Vibe Coding Index05
Coding benchmarks for Gemma 3 27B and GPT-4.1
BenchmarkGemma 3 27BGPT-4.1Leader
Aider polyglot4.9%52.4%GPT-4.1
SWE-bench VerifiedNot measured39.6%Only GPT-4.1 measured
SWE-bench MultimodalNot measured31.1%Only GPT-4.1 measured
AA Coding Index10.1Not measuredOnly Gemma 3 27B measured
LiveCodeBench (Vals AI)Not measured54.7%Only GPT-4.1 measured
LiveCodeBench (Artificial Analysis run)13.7%45.7%GPT-4.1
SciCode23.3%Not measuredOnly Gemma 3 27B measured
Terminal-Bench Hard (Artificial Analysis run)3.8%13.6%GPT-4.1
WeirdMLNot measured39.0%Only GPT-4.1 measured
ALE-BenchNot measured558.1Only GPT-4.1 measured
LMArena Text: coding13221390GPT-4.1
LMArena Text: software and IT13541391GPT-4.1
Neither model has a result on the other 26 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for Gemma 3 27B and GPT-4.1
BenchmarkGemma 3 27BGPT-4.1Leader
AA Intelligence Index4.912.7GPT-4.1
Epoch Capabilities Index130.0136.7GPT-4.1
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
Agentic work benchmarks for Gemma 3 27B and GPT-4.1
BenchmarkGemma 3 27BGPT-4.1Leader
AA Agentic Index0.1Not measuredOnly Gemma 3 27B measured
τ²-Bench Telecom10.5%47.1%GPT-4.1
Neither model has a result on the other 37 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index17
Reasoning benchmarks for Gemma 3 27B and GPT-4.1
BenchmarkGemma 3 27BGPT-4.1Leader
GPQA DiamondNot measured65.4%Only GPT-4.1 measured
MMMU Pro (Vals AI)Not measured72.4%Only GPT-4.1 measured
GPQA Diamond (Artificial Analysis run)42.8%66.6%GPT-4.1
Humanity's Last Exam (Artificial Analysis run)4.4%4.2%Gemma 3 27B
GPQA Diamond (Epoch AI run)47.7%66.9%GPT-4.1
ARC-AGINot measured5.5%Only GPT-4.1 measured
ARC-AGI-2Not measured0.4%Only GPT-4.1 measured
SimpleBenchNot measured27.0%Only GPT-4.1 measured
Chess puzzlesNot measured3.0%Only GPT-4.1 measured
LMArena Text: hard prompts13401384GPT-4.1
LMArena Text: expert13041364GPT-4.1
LMArena Text: hard prompts in English13531393GPT-4.1
LMArena Vision: diagrams11561215GPT-4.1
LMArena Vision: homework11601248GPT-4.1
Neither model has a result on the other 7 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index06
Math benchmarks for Gemma 3 27B and GPT-4.1
BenchmarkGemma 3 27BGPT-4.1Leader
AIMENot measured39.6%Only GPT-4.1 measured
MATH 500 (Vals AI)Not measured87.2%Only GPT-4.1 measured
MGSM (Vals AI)Not measured87.7%Only GPT-4.1 measured
AA Math Index20.734.7GPT-4.1
AIME 2025 (Artificial Analysis run)20.7%34.7%GPT-4.1
OTIS Mock AIME22.5%38.3%GPT-4.1
FrontierMath, tiers 1 to 3Not measured6.0%Only GPT-4.1 measured
MATH level 574.0%83.0%GPT-4.1
LMArena Text: math13121369GPT-4.1
LMArena Text: mathematical work13271360GPT-4.1
Neither model has a result on the other 3 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index24
Knowledge benchmarks for Gemma 3 27B and GPT-4.1
BenchmarkGemma 3 27BGPT-4.1Leader
MMLU ProNot measured80.5%Only GPT-4.1 measured
LegalBench (Vals AI)Not measured83.1%Only GPT-4.1 measured
CaseLaw (Vals AI)Not measured69.9%Only GPT-4.1 measured
CorpFin (Vals AI)Not measured63.0%Only GPT-4.1 measured
TaxEval (Vals AI)Not measured75.1%Only GPT-4.1 measured
MortgageTax (Vals AI)Not measured65.9%Only GPT-4.1 measured
MedQA (Vals AI)Not measured91.2%Only GPT-4.1 measured
MMLU Pro (Artificial Analysis run)66.9%80.6%GPT-4.1
SimpleQA VerifiedNot measured31.1%Only GPT-4.1 measured
GeoBench52.0%72.0%GPT-4.1
LMArena Text: science13831382Gemma 3 27B
LMArena Text: medicine and healthcare13911379Gemma 3 27B
LMArena Text: legal and government13751399GPT-4.1
LMArena Text: business and finance13601369GPT-4.1
Neither model has a result on the other 3 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
Web design benchmarks for Gemma 3 27B and GPT-4.1
BenchmarkGemma 3 27BGPT-4.1Leader
Design Arena: websitesNot measured1039Only GPT-4.1 measured
Design Arena: UI componentsNot measured1000Only GPT-4.1 measured
Design Arena: game devNot measured1080Only GPT-4.1 measured
Design Arena: data vizNot measured1104Only GPT-4.1 measured
Design Arena: 3DNot measured850Only GPT-4.1 measured
Design Arena: all code categoriesNot measured1026Only GPT-4.1 measured
Neither model has a result on the other 3 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index030
General benchmarks for Gemma 3 27B and GPT-4.1
BenchmarkGemma 3 27BGPT-4.1Leader
IFBench31.8%43.0%GPT-4.1
Long context reasoning (AA-LCR)7.3%68.3%GPT-4.1
DTBench52.5%68.3%GPT-4.1
LMCA12.3%25.6%GPT-4.1
Fiction.LiveBench33.3%63.9%GPT-4.1
Creative writing (Lech Mazur)79.9%Not measuredOnly Gemma 3 27B measured
LMArena Text13581383GPT-4.1
LMArena Text: instruction following13211367GPT-4.1
LMArena Text: creative writing13451363GPT-4.1
LMArena Vision11651210GPT-4.1
LMArena Text: English13771392GPT-4.1
LMArena Text: non-English13341370GPT-4.1
LMArena Text: Chinese13461382GPT-4.1
LMArena Text: German13621381GPT-4.1
LMArena Text: Japanese12871319GPT-4.1
LMArena Text: Korean13081338GPT-4.1
LMArena Text: French13681382GPT-4.1
LMArena Text: Spanish13491376GPT-4.1
LMArena Text: Russian13491377GPT-4.1
LMArena Text: Polish13451389GPT-4.1
LMArena Text: multi-turn13451397GPT-4.1
LMArena Text: longer queries13331385GPT-4.1
LMArena Text: writing and language13471371GPT-4.1
LMArena Text: entertainment and media13201365GPT-4.1
LMArena Vision: English11881209GPT-4.1
LMArena Vision: Chinese11641226GPT-4.1
LMArena Vision: reading text11561218GPT-4.1
LMArena Vision: humor11661196GPT-4.1
LMArena Vision: captioning11511207GPT-4.1
LMArena Vision: recognizing things11961208GPT-4.1
LMArena Vision: creative writing11771210GPT-4.1
Neither model has a result on the other 15 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemma 3 27B and GPT-4.1
BenchmarkGemma 3 27BGPT-4.1Leader
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

Gemma 3 27B$0.0003Fits in one request
GPT-4.1$0.0060Fits in one request

Gemma 3 27B costs less

Repository review

50K input + 3K output tokens

Gemma 3 27B$0.0050Fits in one request
GPT-4.1$0.124Fits in one request

Gemma 3 27B costs less

Agent loop

200K input + 10K output tokens

Gemma 3 27B$0.019Exceeds the context window
GPT-4.1$0.480Fits in one request

Gemma 3 27B 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 Gemma 3 27B and GPT-4.1
SpecificationGoogle Gemma 3 27BOpenAI GPT-4.1
MakerGoogleOpenAI
WeightsOpen weightsNo public weights listed
ReleasedMarch 12, 2025April 14, 2025
Context window131K1.05M
Input $/Mtok$0.08$2.00
Output $/Mtok$0.35$8.00
Coding index4.946.5
Independent measurements behind it12
Cost per solved taskNot computable$0.37
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemma 3 27B better than GPT-4.1 for coding?

Pick GPT-4.1 for coding: it scores 46.5 on the Vibe Coding Index against 4.9 for Gemma 3 27B. Pick Gemma 3 27B when output tokens cost 22.9x less, or when what matters is Humanity's Last Exam (Artificial Analysis run) or LMArena Text: science, where it leads.

Which is cheaper, Gemma 3 27B or GPT-4.1?

Gemma 3 27B costs $0.08 per million input tokens and $0.35 per million output tokens; GPT-4.1 costs $2.00 and $8.00. Cost per solved task is known for GPT-4.1 only.

Which has the bigger context window, Gemma 3 27B or GPT-4.1?

GPT-4.1, with 1.05M tokens against 131K.

How many benchmarks is the Gemma 3 27B vs GPT-4.1 comparison based on?

58 benchmarks have a result for both models as of October 21, 2026. Gemma 3 27B's coding index rests on 1 independent benchmark and GPT-4.1's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next