LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

Gemma 4 31B vs GPT-5.4 Mini

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 4 31B Replace GPT-5.4 Mini
Gemma 4 31BGoogle
Coding
38.3
Overall
14.7
Output $/M
$0.40
37
GPT-5.4 MiniOpenAI
Coding
28.2
Overall
24.1
Output $/M
$4.50
21
vs boards won of 58 shared

Decision reading

Pick Gemma 4 31B for coding: it scores 38.3 on the Vibe Coding Index against 28.2 for GPT-5.4 Mini.

Pick GPT-5.4 Mini when the context window is larger, 400K against 262K tokens, or when what matters is LMArena WebDev or AA Coding Index, where it leads.

Model AGemma 4 31BGoogle38.3/1002 independent benchmarksNo cost per solved taskModel BGPT-5.4 MiniOpenAI28.2/1001 independent benchmark$0.70 per solved task
Shared results
58
Gemma 4 31B only
10
GPT-5.4 Mini only
30
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 4 31B 38.3 from 2 benchmarks, GPT-5.4 Mini 28.2 from 1 benchmark.
Widest category
Agentic work, 15.8 points apart on average over 2 shared benchmarks.
Widest single gap
SimpleQA Verified: Gemma 4 31B 10.4% against GPT-5.4 Mini 29.4%, 19.0 points apart on a 0 to 100 scale.
Output price
Gemma 4 31B $0.40 against GPT-5.4 Mini $4.50 per million output tokens, 11.3x apart.
Context window
Gemma 4 31B takes 262K tokens, GPT-5.4 Mini 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
CategoryGemma 4 31BBarsGPT-5.4 MiniTies
Coding28
Intelligence02
Agentic work02
Reasoning53
Math21
Knowledge52
General233
All shared boards3721
CodingLike-for-likeFour of these boards feed the Vibe Coding Index28
Coding benchmarks for Gemma 4 31B and GPT-5.4 Mini
BenchmarkGemma 4 31BGPT-5.4 MiniLeader
LiveBench coding49.0%Not measuredOnly Gemma 4 31B measured
LMArena WebDev13671397GPT-5.4 Mini
AA Coding Index43.456.1GPT-5.4 Mini
LiveCodeBench (Vals AI)Not measured81.5%Only GPT-5.4 Mini measured
SWE-bench Verified (Vals AI)Not measured73.0%Only GPT-5.4 Mini measured
Terminal-Bench 4 (Vals AI)Not measured2.5%Only GPT-5.4 Mini measured
Vibe Code Bench (Vals AI)Not measured48.0%Only GPT-5.4 Mini measured
Code migration (Vals AI)Not measured12.9%Only GPT-5.4 Mini measured
ProgramBench (Vals AI)Not measured0.0%Only GPT-5.4 Mini measured
Terminal-Bench 2.1 (Vals AI)Not measured54.7%Only GPT-5.4 Mini measured
SciCode45.5%52.1%GPT-5.4 Mini
Terminal-Bench Hard (Artificial Analysis run)36.4%52.3%GPT-5.4 Mini
FrontierCodeNot measured27.0%Only GPT-5.4 Mini measured
WeirdML52.3%60.3%GPT-5.4 Mini
ALE-Bench925.51188.6GPT-5.4 Mini
LMArena Text: coding14591438Gemma 4 31B
LMArena Text: software and IT14631436Gemma 4 31B
LMArena WebDev: React13591386GPT-5.4 Mini
LMArena WebDev: HTML13801419GPT-5.4 Mini
Neither model has a result on the other 19 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for Gemma 4 31B and GPT-5.4 Mini
BenchmarkGemma 4 31BGPT-5.4 MiniLeader
AA Intelligence Index14.724.1GPT-5.4 Mini
Epoch Capabilities Index143.0148.7GPT-5.4 Mini
Agentic workLike-for-likePublic boards, shown as published, kept out of the index02
Agentic work benchmarks for Gemma 4 31B and GPT-5.4 Mini
BenchmarkGemma 4 31BGPT-5.4 MiniLeader
AA Agentic Index4.217.9GPT-5.4 Mini
LiveBench research44.4%Not measuredOnly Gemma 4 31B measured
LiveBench browser4.0%Not measuredOnly Gemma 4 31B measured
LiveBench long agentic13.3%Not measuredOnly Gemma 4 31B measured
Finance Agent (Vals AI)Not measured45.4%Only GPT-5.4 Mini measured
Tax Agent Bench (Vals AI)Not measured8.8%Only GPT-5.4 Mini measured
Legal Research Bench (Vals AI)Not measured12.5%Only GPT-5.4 Mini measured
Excel Modeling Benchmark (Vals AI)Not measured45.4%Only GPT-5.4 Mini measured
τ²-Bench Telecom65.5%83.3%GPT-5.4 Mini
DeepResearch BenchNot measured36.3%Only GPT-5.4 Mini measured
Neither model has a result on the other 29 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index53
Reasoning benchmarks for Gemma 4 31B and GPT-5.4 Mini
BenchmarkGemma 4 31BGPT-5.4 MiniLeader
GPQA DiamondNot measured83.1%Only GPT-5.4 Mini measured
LiveBench reasoning68.3%Not measuredOnly Gemma 4 31B measured
MMMU Pro (Vals AI)Not measured79.3%Only GPT-5.4 Mini measured
GPQA Diamond (Artificial Analysis run)85.7%87.5%GPT-5.4 Mini
Humanity's Last Exam (Artificial Analysis run)23.6%28.1%GPT-5.4 Mini
GPQA Diamond (Epoch AI run)75.8%86.9%GPT-5.4 Mini
ARC-AGINot measured63.7%Only GPT-5.4 Mini measured
ARC-AGI-2Not measured18.9%Only GPT-5.4 Mini measured
Chess puzzlesNot measured23.0%Only GPT-5.4 Mini measured
Mystery game puzzlesNot measured11.0%Only GPT-5.4 Mini measured
LMArena Text: hard prompts14481424Gemma 4 31B
LMArena Text: expert14641435Gemma 4 31B
LMArena Text: hard prompts in English14561420Gemma 4 31B
LMArena Vision: diagrams12831266Gemma 4 31B
LMArena Vision: homework13101285Gemma 4 31B
Neither model has a result on the other 6 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index21
Math benchmarks for Gemma 4 31B and GPT-5.4 Mini
BenchmarkGemma 4 31BGPT-5.4 MiniLeader
AIMENot measured95.6%Only GPT-5.4 Mini measured
OTIS Mock AIME73.3%88.9%GPT-5.4 Mini
FrontierMath, tiers 1 to 3Not measured51.2%Only GPT-5.4 Mini measured
FrontierMath, tier 4Not measured9.8%Only GPT-5.4 Mini measured
ProofBenchNot measured21.0%Only GPT-5.4 Mini measured
LMArena Text: math14631418Gemma 4 31B
LMArena Text: mathematical work14651429Gemma 4 31B
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index52
Knowledge benchmarks for Gemma 4 31B and GPT-5.4 Mini
BenchmarkGemma 4 31BGPT-5.4 MiniLeader
MMLU ProNot measured84.5%Only GPT-5.4 Mini measured
CaseLaw (Vals AI)52.6%51.7%Gemma 4 31B
CorpFin (Vals AI)Not measured60.9%Only GPT-5.4 Mini measured
TaxEval (Vals AI)Not measured71.2%Only GPT-5.4 Mini measured
MortgageTax (Vals AI)61.4%63.5%GPT-5.4 Mini
SimpleQA Verified10.4%29.4%GPT-5.4 Mini
LMArena Text: science14571417Gemma 4 31B
LMArena Text: medicine and healthcare14511395Gemma 4 31B
LMArena Text: legal and government14381414Gemma 4 31B
LMArena Text: business and finance14301418Gemma 4 31B
Neither model has a result on the other 7 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
Web design benchmarks for Gemma 4 31B and GPT-5.4 Mini
BenchmarkGemma 4 31BGPT-5.4 MiniLeader
LiveBench design58.4%Not measuredOnly Gemma 4 31B measured
Neither model has a result on the other 8 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index233
General benchmarks for Gemma 4 31B and GPT-5.4 Mini
BenchmarkGemma 4 31BGPT-5.4 MiniLeader
LiveBench data analysis53.3%Not measuredOnly Gemma 4 31B measured
SAGE (Vals AI)55.0%50.8%Gemma 4 31B
Vals Multimodal IndexNot measured54.2%Only GPT-5.4 Mini measured
IFBench75.6%73.3%Gemma 4 31B
Long context reasoning (AA-LCR)69.7%77.0%GPT-5.4 Mini
DTBench82.7%80.0%Gemma 4 31B
LMCA39.3%40.8%GPT-5.4 Mini
Surface Evolver Bench30.6%Not measuredOnly Gemma 4 31B measured
LMArena Text14431412Gemma 4 31B
LMArena Text: instruction following14331405Gemma 4 31B
LMArena Text: creative writing14171370Gemma 4 31B
LMArena Vision12761245Gemma 4 31B
LMArena Text: English14511412Gemma 4 31B
LMArena Text: non-English14301405Gemma 4 31B
LMArena Text: Chinese14751446Gemma 4 31B
LMArena Text: GermanNot measured1408Only GPT-5.4 Mini measured
LMArena Text: JapaneseNot measured1375Only GPT-5.4 Mini measured
LMArena Text: KoreanNot measured1370Only GPT-5.4 Mini measured
LMArena Text: French14361441GPT-5.4 Mini
LMArena Text: Spanish14421405Gemma 4 31B
LMArena Text: Russian14601417Gemma 4 31B
LMArena Text: Polish14331415Gemma 4 31B
LMArena Text: multi-turn14521429Gemma 4 31B
LMArena Text: longer queries14461408Gemma 4 31B
LMArena Text: writing and language14321393Gemma 4 31B
LMArena Text: entertainment and media14051373Gemma 4 31B
LMArena Vision: English12711244Gemma 4 31B
LMArena Vision: Chinese13261286Gemma 4 31B
LMArena Vision: reading text12821253Gemma 4 31B
LMArena Vision: humor12801235Gemma 4 31B
LMArena Vision: captioning1218Not measuredOnly Gemma 4 31B measured
LMArena Vision: recognizing things12261218Gemma 4 31B
LMArena Vision: creative writing12851222Gemma 4 31B
LMArena Document1425Not measuredOnly Gemma 4 31B measured
Neither model has a result on the other 12 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemma 4 31B and GPT-5.4 Mini
BenchmarkGemma 4 31BGPT-5.4 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

Gemma 4 31B$0.0004Fits in one request
GPT-5.4 Mini$0.0030Fits in one request

Gemma 4 31B costs less

Repository review

50K input + 3K output tokens

Gemma 4 31B$0.011Fits in one request
GPT-5.4 Mini$0.051Fits in one request

Gemma 4 31B costs less

Agent loop

200K input + 10K output tokens

Gemma 4 31B$0.044Fits in one request
GPT-5.4 Mini$0.195Fits in one request

Gemma 4 31B 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 4 31B and GPT-5.4 Mini
SpecificationGoogle Gemma 4 31BOpenAI GPT-5.4 Mini
MakerGoogleOpenAI
WeightsOpen weightsNo public weights listed
ReleasedApril 2, 2026March 17, 2026
Context window262K400K
Input $/Mtok$0.20$0.75
Output $/Mtok$0.40$4.50
Coding index38.328.2
Independent measurements behind it21
Cost per solved taskNot computable$0.70
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemma 4 31B better than GPT-5.4 Mini for coding?

Pick Gemma 4 31B for coding: it scores 38.3 on the Vibe Coding Index against 28.2 for GPT-5.4 Mini. Pick GPT-5.4 Mini when the context window is larger, 400K against 262K tokens, or when what matters is LMArena WebDev or AA Coding Index, where it leads.

Which is cheaper, Gemma 4 31B or GPT-5.4 Mini?

Gemma 4 31B costs $0.20 per million input tokens and $0.40 per million output tokens; GPT-5.4 Mini costs $0.75 and $4.50. Cost per solved task is known for GPT-5.4 Mini only.

Which has the bigger context window, Gemma 4 31B or GPT-5.4 Mini?

GPT-5.4 Mini, with 400K tokens against 262K.

How many benchmarks is the Gemma 4 31B vs GPT-5.4 Mini comparison based on?

58 benchmarks have a result for both models as of October 21, 2026. Gemma 4 31B's coding index rests on 2 independent benchmarks and GPT-5.4 Mini's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next