Model comparison

GPT-5.5 vs Llama 4 Maverick

GPT-5.5 is the stronger model overall, scoring 63.4 to 30.6 on the Noometry Index. Llama 4 Maverick costs 37× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

Last verified October 10, 2026. 36 shared benchmarks.

GPT-5.5 OpenAI

63.4

Rank #8 Confirmed

Llama 4 Maverick

Summary

They share 36 benchmarks with published results for both. GPT-5.5 scores higher in 10 categories and Llama 4 Maverick in 0 categories; 10 gaps are clear of the uncertainty.
The widest gap is in reasoning, where GPT-5.5 leads 72.7 to 10.1.
The biggest single-benchmark swing is ARC-AGI-1: 95% for GPT-5.5 and 4.4% for Llama 4 Maverick.
Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $5 / $30 for GPT-5.5.
GPT-5.5 accepts more context: 1.05M tokens versus 128K.
Llama 4 Maverick has downloadable open weights; the other is API-only.

Side by side

GPT-5.5 and Llama 4 Maverick specifications
	GPT-5.5	Llama 4 Maverick
Provider	OpenAI	Meta
Noometry Index	63.4	30.6
Released	2026-04-23	2025-04-05
Weights	Proprietary	Open
Context window	1.05M	128K
Max output	128K	4K
Input $ / M tokens	$5	$0.19
Output $ / M tokens	$30	$0.65
Results tracked	71	54

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding GPT-5.5 leads

GPT-5.5: 58.3 (#13), Llama 4 Maverick: 26.3 (#318)

Coding benchmarks
Benchmark	GPT-5.5	Llama 4 Maverick
SciCode	56.1%	33.1%
WeirdML	84.9%	24.5%
LMArena Coding	1494	1302
ALE-Bench	1,943	172.97
SWE-bench Verified	80.6%	—
DeepSWE	67%	—
FrontierCode	43%	—
SWE-bench Verified (bash only)	—	21%
Aider Polyglot	—	15.6%
LMArena WebDev	1512	—
GSO	40.2%	—
BigCodeBench Instruct	—	49.7%
MirrorCode	10%	—
BigCodeBench Complete	—	61.4%

Agentic & Tool Use GPT-5.5 leads

GPT-5.5: 50.7 (#6), Llama 4 Maverick: 27.9 (#86)

Agentic & Tool Use benchmarks
Benchmark	GPT-5.5	Llama 4 Maverick
Terminal-Bench	84.7%	—
APEX-Agents	55.1%	—
Berkeley Function Calling Leaderboard	—	37.3%
OSWorld 2.0	13%	—
Remote Labor Index	6.3%	—
τ²-bench Banking	44.6%	—
DeepResearch Bench	54%	—
PostTrainBench	27.2%	—
ExploitBench	47.4%	—
GBAEval	53.2%	—
GDP.pdf	26%	—
LMArena Search	1242	—
Vending-Bench 2	7,524	—

Reasoning GPT-5.5 leads

GPT-5.5: 72.7 (#8), Llama 4 Maverick: 10.1 (#337)

Reasoning benchmarks
Benchmark	GPT-5.5	Llama 4 Maverick
ARC-AGI-2	85%	0%
SimpleBench	69%	27.7%
Kagi LLM Benchmark	88.8%	55.9%
NYT Connections (extended)	96.2%	8%
ARC-AGI-1	95%	4.4%
CritPt	27.1%	0%
LMArena Hard Prompts	1489	1281
DTBench	96%	61.9%
LMCA	54.3%	15.9%
Epoch Capabilities Index	159.25	132.2
ForecastBench	60.6	57.5
Chess Puzzles	54%	—
EnigmaEval	—	0.6%
EBR-Bench	34.3%	—
Mystery Game Puzzles	56%	—
Surface Evolver Bench	88.1%	—
Bench to the Future 3	0.14	—

Math GPT-5.5 leads

GPT-5.5: 81.5 (#9), Llama 4 Maverick: 24.6 (#256)

Math benchmarks
Benchmark	GPT-5.5	Llama 4 Maverick
OTIS Mock AIME 2024-2025	100%	20.6%
LMArena Math	1486	1299
FrontierMath (Feb 2025 set)	51.7%	0.7%
FrontierMath (Tiers 1-3)	85.3%	—
FrontierMath Tier 4	72.5%	—
MathArena Final-Answer Competitions	94.3%	—
ProofBench	50%	—
Omni-MATH	—	42.2%
MATH Level 5	—	73%
FrontierMath Erdős	0%	—
FrontierMath Tier 4 (v1)	35.4%	—

Knowledge GPT-5.5 leads

GPT-5.5: 64.4 (#14), Llama 4 Maverick: 33.3 (#196)

Knowledge benchmarks
Benchmark	GPT-5.5	Llama 4 Maverick
GPQA Diamond	94%	67%
Vectara Hallucination Rate	9.3%	8.2%
LMArena Expert	1508	1259
Humanity's Last Exam	—	5.7%
SimpleQA Verified	63%	—
MMLU-Pro	—	81%
Confabulations	—	22.6%
GPQA (HELM)	—	65%

Multimodal GPT-5.5 leads

GPT-5.5: 47.1 (#11), Llama 4 Maverick: 30.9 (#100)

Multimodal benchmarks
Benchmark	GPT-5.5	Llama 4 Maverick
LMArena Vision	1297	1141
GeoBench	—	52%
Blueprint-Bench 2	36.2%	—
Furniture Assembly	44.2%	—
LMArena Document	1486	—
SpatialViz-Bench	—	31.8%

Multilingual GPT-5.5 leads

GPT-5.5: 56.5 (#19), Llama 4 Maverick: 42.4 (#190)

Multilingual benchmarks
Benchmark	GPT-5.5	Llama 4 Maverick
LMArena Non-English	1467	1269
LMArena Chinese	1535	1277
LMArena French	1486	1258
LMArena German	1478	1291
LMArena Japanese	1498	1207
LMArena Korean	1461	1203
LMArena Russian	1473	1286
LMArena Spanish	1469	1293

Instruction Following GPT-5.5 leads

GPT-5.5: 77.7 (#18), Llama 4 Maverick: 71.9 (#140)

Instruction Following benchmarks
Benchmark	GPT-5.5	Llama 4 Maverick
LMArena Instruction Following	1479	1267
IFEval	—	90.8%

Long Context GPT-5.5 leads

GPT-5.5: 48.3 (#12), Llama 4 Maverick: 31.4 (#274)

Long Context benchmarks
Benchmark	GPT-5.5	Llama 4 Maverick
LMArena Longer Query	1484	1280
Fiction.LiveBench	—	46.2%
CL-bench Life	22.2%	—

Writing & Preference GPT-5.5 leads

GPT-5.5: 72.8 (#13), Llama 4 Maverick: 39.0 (#247)

Writing & Preference benchmarks
Benchmark	GPT-5.5	Llama 4 Maverick
LMArena Text	1471	1287
LMArena Creative Writing	1455	1266
EQ-Bench Creative Writing	1844	860
LMArena Multi-Turn	1477	1289
Short-Story Creative Writing	—	62%
WildBench	—	80%
EQ-Bench 4	1315	—

Frequently asked questions

Is GPT-5.5 better than Llama 4 Maverick?

Which is cheaper, GPT-5.5 or Llama 4 Maverick?

Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; GPT-5.5 lists at $5 and $30.

Is GPT-5.5 or Llama 4 Maverick better for coding?

GPT-5.5 scores higher on coding benchmarks: 58.3 versus 26.3 in the Noometry coding category.

Which has the bigger context window?

GPT-5.5 does, with 1.05M tokens against 128K.

How many benchmarks do GPT-5.5 and Llama 4 Maverick share?

36 benchmarks have published results for both models. GPT-5.5 has 71 scored results on Noometry and Llama 4 Maverick has 54.