LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

GPT-5.4 vs Llama 4 Maverick

Updated October 21, 2026. 70 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GPT-5.4 Replace Llama 4 Maverick
GPT-5.4OpenAI
Coding
37.7
Overall
39.0
Output $/M
$15.00
70
Llama 4 MaverickMeta Llama
Coding
18.1
Overall
9.3
Output $/M
$0.6525
0
vs boards won of 70 shared

Decision reading

Pick GPT-5.4 for coding: it scores 37.7 on the Vibe Coding Index against 18.1 for Llama 4 Maverick, though on 1 benchmark against 2.

Pick Llama 4 Maverick when output tokens cost 23.0x less.

Model AGPT-5.4OpenAI37.7/1001 independent benchmark$1.02 per solved taskModel BLlama 4 MaverickMeta Llama18.1/1002 independent benchmarks$1.491 per solved task
Shared results
70
GPT-5.4 only
39
Llama 4 Maverick only
14
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-5.4 37.7 from 1 benchmark, Llama 4 Maverick 18.1 from 2 benchmarks.
Widest category
Agentic work, 69.3 points apart on average over 1 shared benchmark.
Widest single gap
ARC-AGI: GPT-5.4 93.7% against Llama 4 Maverick 4.4%, 89.3 points apart on a 0 to 100 scale.
Output price
GPT-5.4 $15.00 against Llama 4 Maverick $0.6525 per million output tokens, 23.0x apart.
Context window
GPT-5.4 takes 1.05M tokens, Llama 4 Maverick 524K.

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-5.4BarsLlama 4 MaverickTies
Coding70
Intelligence20
Agentic work10
Reasoning110
Math30
Knowledge110
Web design60
General290
All shared boards700
CodingLike-for-likeFour of these boards feed the Vibe Coding Index70
Coding benchmarks for GPT-5.4 and Llama 4 Maverick
BenchmarkGPT-5.4Llama 4 MaverickLeader
Aider polyglotNot measured15.6%Only Llama 4 Maverick measured
SWE-bench VerifiedNot measured21.0%Only Llama 4 Maverick measured
LMArena WebDev1464Not measuredOnly GPT-5.4 measured
AA Coding Index71.116.3GPT-5.4
LiveCodeBench (Vals AI)84.1%47.3%GPT-5.4
SWE-bench Verified (Vals AI)78.2%Not measuredOnly GPT-5.4 measured
Vibe Code Bench (Vals AI)67.4%Not measuredOnly GPT-5.4 measured
Code migration (Vals AI)35.0%Not measuredOnly GPT-5.4 measured
ProgramBench (Vals AI)0.5%Not measuredOnly GPT-5.4 measured
SkillsBench (Vals AI)51.7%Not measuredOnly GPT-5.4 measured
LiveCodeBench (Artificial Analysis run)Not measured39.7%Only Llama 4 Maverick measured
SciCodeNot measured31.7%Only Llama 4 Maverick measured
Terminal-Bench Hard (Artificial Analysis run)57.6%6.8%GPT-5.4
Terminal-Bench81.8%Not measuredOnly GPT-5.4 measured
GSO-Bench31.4%Not measuredOnly GPT-5.4 measured
WeirdML77.7%24.5%GPT-5.4
MirrorCode15.6%Not measuredOnly GPT-5.4 measured
ALE-Bench1607.0173.0GPT-5.4
LMArena Text: coding14961301GPT-5.4
LMArena Text: software and IT14911299GPT-5.4
LMArena WebDev: React1454Not measuredOnly GPT-5.4 measured
LMArena WebDev: HTML1478Not measuredOnly GPT-5.4 measured
LMArena WebDev: from an image1436Not measuredOnly GPT-5.4 measured
Neither model has a result on the other 15 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for GPT-5.4 and Llama 4 Maverick
BenchmarkGPT-5.4Llama 4 MaverickLeader
AA Intelligence Index39.09.3GPT-5.4
Epoch Capabilities Index157.3132.2GPT-5.4
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for GPT-5.4 and Llama 4 Maverick
BenchmarkGPT-5.4Llama 4 MaverickLeader
AA Agentic IndexNot measured0.6Only Llama 4 Maverick measured
Design Arena: agents building web apps1059Not measuredOnly GPT-5.4 measured
Design Arena: agents building full-stack apps1017Not measuredOnly GPT-5.4 measured
Design Arena: agents building mobile apps1087Not measuredOnly GPT-5.4 measured
Design Arena: agents building Android apps1037Not measuredOnly GPT-5.4 measured
Design Arena: agents building Godot games1135Not measuredOnly GPT-5.4 measured
τ²-Bench Telecom87.1%17.8%GPT-5.4
APEX-Agents52.4%Not measuredOnly GPT-5.4 measured
METR time horizons74.3%Not measuredOnly GPT-5.4 measured
DeepResearch Bench35.1%Not measuredOnly GPT-5.4 measured
PostTrainBench19.0%Not measuredOnly GPT-5.4 measured
Vending-Bench 2$6,144Not measuredOnly GPT-5.4 measured
Neither model has a result on the other 27 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index110
Reasoning benchmarks for GPT-5.4 and Llama 4 Maverick
BenchmarkGPT-5.4Llama 4 MaverickLeader
GPQA Diamond91.7%69.4%GPT-5.4
MMMU Pro (Vals AI)87.5%71.7%GPT-5.4
GPQA Diamond (Artificial Analysis run)92.0%67.1%GPT-5.4
Humanity's Last Exam (Artificial Analysis run)43.7%4.9%GPT-5.4
GPQA Diamond (Epoch AI run)93.3%Not measuredOnly GPT-5.4 measured
ARC-AGI93.7%4.4%GPT-5.4
ARC-AGI-274.0%0.0%GPT-5.4
SimpleBenchNot measured27.7%Only Llama 4 Maverick measured
Chess puzzles48.0%Not measuredOnly GPT-5.4 measured
Mystery game puzzles37.0%Not measuredOnly GPT-5.4 measured
EBR-bench25.4%Not measuredOnly GPT-5.4 measured
LMArena Text: hard prompts14851281GPT-5.4
LMArena Text: expert15071259GPT-5.4
LMArena Text: hard prompts in English14791295GPT-5.4
LMArena Vision: diagrams13271134GPT-5.4
LMArena Vision: homework13361164GPT-5.4
Neither model has a result on the other 5 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index30
Math benchmarks for GPT-5.4 and Llama 4 Maverick
BenchmarkGPT-5.4Llama 4 MaverickLeader
AIME96.7%25.2%GPT-5.4
MATH 500 (Vals AI)Not measured85.2%Only Llama 4 Maverick measured
MGSM (Vals AI)Not measured92.4%Only Llama 4 Maverick measured
AA Math IndexNot measured19.3Only Llama 4 Maverick measured
AIME 2025 (Artificial Analysis run)Not measured19.3%Only Llama 4 Maverick measured
OTIS Mock AIME97.8%Not measuredOnly GPT-5.4 measured
FrontierMath, tiers 1 to 378.6%Not measuredOnly GPT-5.4 measured
FrontierMath, tier 449.0%Not measuredOnly GPT-5.4 measured
ProofBench56.0%Not measuredOnly GPT-5.4 measured
LMArena Text: math14881299GPT-5.4
LMArena Text: mathematical work14971295GPT-5.4
Neither model has a result on the other 2 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index110
Knowledge benchmarks for GPT-5.4 and Llama 4 Maverick
BenchmarkGPT-5.4Llama 4 MaverickLeader
MMLU Pro87.5%79.4%GPT-5.4
LegalBench (Vals AI)86.0%77.8%GPT-5.4
CaseLaw (Vals AI)63.8%Not measuredOnly GPT-5.4 measured
CorpFin (Vals AI)65.3%49.7%GPT-5.4
TaxEval (Vals AI)74.0%66.6%GPT-5.4
MortgageTax (Vals AI)68.3%58.5%GPT-5.4
MedQA (Vals AI)96.1%43.3%GPT-5.4
MedCode (Vals AI)41.3%36.5%GPT-5.4
MMLU Pro (Artificial Analysis run)Not measured80.9%Only Llama 4 Maverick measured
SimpleQA Verified45.1%Not measuredOnly GPT-5.4 measured
GeoBenchNot measured52.0%Only Llama 4 Maverick measured
LMArena Text: science14791288GPT-5.4
LMArena Text: medicine and healthcare14581275GPT-5.4
LMArena Text: legal and government14861286GPT-5.4
LMArena Text: business and finance14741260GPT-5.4
Neither model has a result on the other 2 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index60
Web design benchmarks for GPT-5.4 and Llama 4 Maverick
BenchmarkGPT-5.4Llama 4 MaverickLeader
Design Arena: websites1199871GPT-5.4
Design Arena: UI components1215899GPT-5.4
Design Arena: game dev1195838GPT-5.4
Design Arena: data viz1196882GPT-5.4
Design Arena: 3D1065901GPT-5.4
Design Arena: all code categories1180880GPT-5.4
Design Arena: SVG1141Not measuredOnly GPT-5.4 measured
Design Arena: ASCII art1188Not measuredOnly GPT-5.4 measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index290
General benchmarks for GPT-5.4 and Llama 4 Maverick
BenchmarkGPT-5.4Llama 4 MaverickLeader
SAGE (Vals AI)43.3%35.7%GPT-5.4
MedScribe (Vals AI)77.5%54.2%GPT-5.4
IFBench73.9%43.0%GPT-5.4
Long context reasoning (AA-LCR)82.0%50.0%GPT-5.4
DTBench94.4%61.9%GPT-5.4
LMCA52.0%15.9%GPT-5.4
Fiction.LiveBenchNot measured46.2%Only Llama 4 Maverick measured
Creative writing (Lech Mazur)Not measured62.0%Only Llama 4 Maverick measured
CL-bench27.9%Not measuredOnly GPT-5.4 measured
Furniture assembly37.5%Not measuredOnly GPT-5.4 measured
CL-bench Life21.7%Not measuredOnly GPT-5.4 measured
LMArena Text14691287GPT-5.4
LMArena Text: instruction following14691267GPT-5.4
LMArena Text: creative writing14391266GPT-5.4
LMArena Vision13031142GPT-5.4
LMArena Text: English14671303GPT-5.4
LMArena Text: non-English14651269GPT-5.4
LMArena Text: Chinese15201277GPT-5.4
LMArena Text: German14721291GPT-5.4
LMArena Text: Japanese14851206GPT-5.4
LMArena Text: Korean14491203GPT-5.4
LMArena Text: French14931259GPT-5.4
LMArena Text: Spanish14521293GPT-5.4
LMArena Text: Russian14801286GPT-5.4
LMArena Text: Polish14721275GPT-5.4
LMArena Text: multi-turn14821289GPT-5.4
LMArena Text: longer queries14731280GPT-5.4
LMArena Text: writing and language14581277GPT-5.4
LMArena Text: entertainment and media14381250GPT-5.4
LMArena Vision: English13011150GPT-5.4
LMArena Vision: Chinese13541141GPT-5.4
LMArena Vision: reading text13121143GPT-5.4
LMArena Vision: humor12841100GPT-5.4
LMArena Vision: recognizing things1251Not measuredOnly GPT-5.4 measured
LMArena Vision: creative writing12831099GPT-5.4
LMArena Document1471Not measuredOnly GPT-5.4 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-5.4 and Llama 4 Maverick
BenchmarkGPT-5.4Llama 4 MaverickLeader
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-5.4$0.010Fits in one request
Llama 4 Maverick$0.0005Fits in one request

Llama 4 Maverick costs less

Repository review

50K input + 3K output tokens

GPT-5.4$0.170Fits in one request
Llama 4 Maverick$0.011Fits in one request

Llama 4 Maverick costs less

Agent loop

200K input + 10K output tokens

GPT-5.4$0.650Fits in one request
Llama 4 Maverick$0.044Fits in one request

Llama 4 Maverick 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-5.4 and Llama 4 Maverick
SpecificationOpenAI GPT-5.4Meta Llama 4 Maverick
MakerOpenAIMeta Llama
WeightsNo public weights listedOpen weights
ReleasedMarch 5, 2026April 5, 2025
Context window1.05M524K
Input $/Mtok$2.50$0.1875
Output $/Mtok$15.00$0.6525
Coding index37.718.1
Independent measurements behind it12
Cost per solved task$1.02$1.491
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-5.4 better than Llama 4 Maverick for coding?

Pick GPT-5.4 for coding: it scores 37.7 on the Vibe Coding Index against 18.1 for Llama 4 Maverick, though on 1 benchmark against 2. Pick Llama 4 Maverick when output tokens cost 23.0x less.

Which is cheaper, GPT-5.4 or Llama 4 Maverick?

GPT-5.4 costs $2.50 per million input tokens and $15.00 per million output tokens; Llama 4 Maverick costs $0.1875 and $0.6525. Per solved SWE-bench Verified task, GPT-5.4 costs $1.02 and Llama 4 Maverick $1.491.

Which has the bigger context window, GPT-5.4 or Llama 4 Maverick?

GPT-5.4, with 1.05M tokens against 524K.

How many benchmarks is the GPT-5.4 vs Llama 4 Maverick comparison based on?

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

Compare next