Model comparison

GPT-5.6 Sol vs Llama 4 Maverick

GPT-5.6 Sol is the stronger model overall, scoring 65.0 to 30.9 on the Noometry Index. Llama 4 Maverick costs 26× less per token, which makes it the better buy when GPT-5.6 Sol's lead doesn't matter for your workload.

Last verified October 10, 2026. 35 shared benchmarks.

GPT-5.6 Sol OpenAI

65.0

Rank #7 Confirmed

Llama 4 Maverick

Summary

They share 35 benchmarks with published results for both. GPT-5.6 Sol 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-5.6 Sol leads 74.8 to 10.1.
The biggest single-benchmark swing is ARC-AGI-1: 97.5% for GPT-5.6 Sol and 4.4% for Llama 4 Maverick.
Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $4 / $20 for GPT-5.6 Sol.
GPT-5.6 Sol accepts more context: 1.05M tokens versus 128K.
Llama 4 Maverick has downloadable open weights; the other is API-only.

Side by side

GPT-5.6 Sol and Llama 4 Maverick specifications
	GPT-5.6 Sol	Llama 4 Maverick
Provider	OpenAI	Meta
Noometry Index	65.0	30.9
Released	2026-07-09	2025-04-05
Weights	Proprietary	Open
Context window	1.05M	128K
Max output	128K	4K
Input $ / M tokens	$4	$0.19
Output $ / M tokens	$20	$0.65
Results tracked	65	54

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

Category by category

Coding GPT-5.6 Sol leads

GPT-5.6 Sol: 65.1 (#7), Llama 4 Maverick: 26.6 (#324)

Coding benchmarks
Benchmark	GPT-5.6 Sol	Llama 4 Maverick
SciCode	57.1%	33.1%
WeirdML	89.4%	24.5%
LMArena Coding	1498	1302
ALE-Bench	2,177	172.97
DeepSWE	72.7%	—
FrontierCode	47.5%	—
SWE-bench Verified (bash only)	—	21%
Aider Polyglot	—	15.6%
CursorBench	41.7%	—
LMArena WebDev	1618	—
FrontierSWE	32.2%	—
GSO	76.5%	—
BigCodeBench Instruct	—	49.7%
MirrorCode	20%	—
BigCodeBench Complete	—	61.4%

Agentic & Tool Use GPT-5.6 Sol leads

GPT-5.6 Sol: 50.3 (#7), Llama 4 Maverick: 28.2 (#91)

Agentic & Tool Use benchmarks
Benchmark	GPT-5.6 Sol	Llama 4 Maverick
APEX-Agents	51.4%	—
Berkeley Function Calling Leaderboard	—	37.3%
OSWorld 2.0	27.3%	—
τ²-bench Banking	46.9%	—
PostTrainBench	36.2%	—
BALROG	60%	—
GBAEval	52.6%	—
GDP.pdf	30.7%	—
LMArena Search	1257	—
Vending-Bench 2	9,619	—

Reasoning GPT-5.6 Sol leads

GPT-5.6 Sol: 74.8 (#8), Llama 4 Maverick: 10.1 (#342)

Reasoning benchmarks
Benchmark	GPT-5.6 Sol	Llama 4 Maverick
ARC-AGI-2	92.5%	0%
SimpleBench	71.7%	27.7%
Kagi LLM Benchmark	67%	55.9%
NYT Connections (extended)	93.8%	8%
ARC-AGI-1	97.5%	4.4%
CritPt	32.3%	0%
EnigmaEval	37.1%	0.6%
LMArena Hard Prompts	1484	1281
DTBench	96%	61.9%
LMCA	59.2%	15.9%
Epoch Capabilities Index	161.66	132.2
Chess Puzzles	64%	—
EBR-Bench	44.8%	—
Mystery Game Puzzles	58%	—
Surface Evolver Bench	93.1%	—
Bench to the Future 3	0.14	—
ForecastBench	—	57.5

Math GPT-5.6 Sol leads

GPT-5.6 Sol: 85.6 (#9), Llama 4 Maverick: 26.0 (#262)

Math benchmarks
Benchmark	GPT-5.6 Sol	Llama 4 Maverick
OTIS Mock AIME 2024-2025	100%	20.6%
LMArena Math	1474	1299
FrontierMath (Tiers 1-3)	89.1%	—
FrontierMath Tier 4	82.9%	—
ProofBench	83%	—
Omni-MATH	—	42.2%
MATH Level 5	—	73%
FrontierMath (Feb 2025 set)	—	0.7%
FrontierMath Erdős	0%	—

Knowledge GPT-5.6 Sol leads

GPT-5.6 Sol: 64.3 (#18), Llama 4 Maverick: 33.4 (#204)

Knowledge benchmarks
Benchmark	GPT-5.6 Sol	Llama 4 Maverick
GPQA Diamond	93.5%	67%
Vectara Hallucination Rate	12.4%	8.2%
LMArena Expert	1516	1259
Humanity's Last Exam	—	5.7%
SimpleQA Verified	69.7%	—
MMLU-Pro	—	81%
Confabulations	—	22.6%
GPQA (HELM)	—	65%

Multimodal GPT-5.6 Sol leads

GPT-5.6 Sol: 48.6 (#9), Llama 4 Maverick: 31.6 (#105)

Multimodal benchmarks
Benchmark	GPT-5.6 Sol	Llama 4 Maverick
LMArena Vision	1281	1142
GeoBench	—	52%
Blueprint-Bench 2	33.6%	—
Furniture Assembly	56.7%	—
LMArena Document	1483	—
SpatialViz-Bench	—	31.8%

Multilingual GPT-5.6 Sol leads

GPT-5.6 Sol: 55.3 (#32), Llama 4 Maverick: 42.2 (#195)

Multilingual benchmarks
Benchmark	GPT-5.6 Sol	Llama 4 Maverick
LMArena Non-English	1452	1269
LMArena Chinese	1527	1277
LMArena French	1477	1259
LMArena German	1476	1291
LMArena Japanese	1471	1207
LMArena Korean	1442	1203
LMArena Russian	1468	1286
LMArena Spanish	1441	1293

Instruction Following GPT-5.6 Sol leads

GPT-5.6 Sol: 77.7 (#16), Llama 4 Maverick: 71.7 (#146)

Instruction Following benchmarks
Benchmark	GPT-5.6 Sol	Llama 4 Maverick
LMArena Instruction Following	1482	1267
IFEval	—	90.8%

Long Context GPT-5.6 Sol leads

GPT-5.6 Sol: 45.4 (#42), Llama 4 Maverick: 31.4 (#279)

Long Context benchmarks
Benchmark	GPT-5.6 Sol	Llama 4 Maverick
LMArena Longer Query	1480	1280
Fiction.LiveBench	—	46.2%

Writing & Preference GPT-5.6 Sol leads

GPT-5.6 Sol: 73.3 (#12), Llama 4 Maverick: 38.8 (#252)

Writing & Preference benchmarks
Benchmark	GPT-5.6 Sol	Llama 4 Maverick
LMArena Text	1457	1287
LMArena Creative Writing	1448	1267
EQ-Bench Creative Writing	1972	860
LMArena Multi-Turn	1460	1289
Short-Story Creative Writing	—	62%
WildBench	—	80%
EQ-Bench 4	1250	—

Frequently asked questions

Is GPT-5.6 Sol better than Llama 4 Maverick?

Which is cheaper, GPT-5.6 Sol 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-5.6 Sol lists at $4 and $20.

Is GPT-5.6 Sol or Llama 4 Maverick better for coding?

GPT-5.6 Sol scores higher on coding benchmarks: 65.1 versus 26.6 in the Noometry coding category.

Which has the bigger context window?

GPT-5.6 Sol does, with 1.05M tokens against 128K.

How many benchmarks do GPT-5.6 Sol and Llama 4 Maverick share?

35 benchmarks have published results for both models. GPT-5.6 Sol has 65 scored results on Noometry and Llama 4 Maverick has 54.