Noometry

Model comparison

GPT-5.6 Terra vs Llama 4 Maverick

GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 30.9 on the Noometry Index. Llama 4 Maverick costs 15× less per token, which makes it the better buy when GPT-5.6 Terra's lead doesn't matter for your workload.

Last verified . 33 shared benchmarks.

GPT-5.6 Terra OpenAI

59.2

Rank #17 Confirmed

Llama 4 Maverick Meta

30.9

Rank #282 Confirmed

Summary

  • They share 33 benchmarks with published results for both. GPT-5.6 Terra scores higher in 10 categories and Llama 4 Maverick in 0 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-5.6 Terra leads 81.6 to 26.0.
  • The biggest single-benchmark swing is ARC-AGI-1: 96.5% for GPT-5.6 Terra and 4.4% for Llama 4 Maverick.
  • Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $2 / $12 for GPT-5.6 Terra.
  • GPT-5.6 Terra 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 TerraLlama 4 Maverick
ProviderOpenAIMeta
Noometry Index59.230.9
Released2026-07-092025-04-05
WeightsProprietaryOpen
Context window1.05M128K
Max output128K4K
Input $ / M tokens$2$0.19
Output $ / M tokens$12$0.65
Results tracked5254

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

Category by category

Coding GPT-5.6 Terra leads

GPT-5.6 Terra: 57.7 (#19), Llama 4 Maverick: 26.6 (#324)

BenchmarkGPT-5.6 TerraLlama 4 Maverick
SciCode55%33.1%
WeirdML78.3%24.5%
LMArena Coding14841302
ALE-Bench1,951172.97
DeepSWE69.6%—
FrontierCode41.3%—
SWE-bench Verified (bash only)—21%
Aider Polyglot—15.6%
CursorBench41.3%—
LMArena WebDev1522—
BigCodeBench Instruct—49.7%
BigCodeBench Complete—61.4%

Agentic & Tool Use GPT-5.6 Terra leads

GPT-5.6 Terra: 40.1 (#25), Llama 4 Maverick: 28.2 (#91)

BenchmarkGPT-5.6 TerraLlama 4 Maverick
APEX-Agents58.2%—
Berkeley Function Calling Leaderboard—37.3%
BALROG53.2%—
GDP.pdf24.7%—
Vending-Bench 27,343—

Reasoning GPT-5.6 Terra leads

GPT-5.6 Terra: 60.7 (#21), Llama 4 Maverick: 10.1 (#342)

BenchmarkGPT-5.6 TerraLlama 4 Maverick
ARC-AGI-283.9%0%
SimpleBench48.9%27.7%
Kagi LLM Benchmark51.3%55.9%
NYT Connections (extended)78.4%8%
ARC-AGI-196.5%4.4%
CritPt30%0%
LMArena Hard Prompts14681281
DTBench93.3%61.9%
LMCA55%15.9%
Epoch Capabilities Index159.62132.2
Chess Puzzles54%—
EnigmaEval—0.6%
Mystery Game Puzzles35%—
Surface Evolver Bench83.8%—
ForecastBench—57.5

Math GPT-5.6 Terra leads

GPT-5.6 Terra: 81.6 (#12), Llama 4 Maverick: 26.0 (#262)

BenchmarkGPT-5.6 TerraLlama 4 Maverick
OTIS Mock AIME 2024-202599.7%20.6%
LMArena Math14661299
FrontierMath (Tiers 1-3)86%—
FrontierMath Tier 470.7%—
ProofBench74%—
Omni-MATH—42.2%
MATH Level 5—73%
FrontierMath (Feb 2025 set)—0.7%

Knowledge GPT-5.6 Terra leads

GPT-5.6 Terra: 61.2 (#30), Llama 4 Maverick: 33.4 (#204)

BenchmarkGPT-5.6 TerraLlama 4 Maverick
GPQA Diamond93.3%67%
LMArena Expert14921259
Humanity's Last Exam—5.7%
SimpleQA Verified43.2%—
MMLU-Pro—81%
Confabulations—22.6%
Vectara Hallucination Rate—8.2%
GPQA (HELM)—65%

Multimodal GPT-5.6 Terra leads

GPT-5.6 Terra: 47.3 (#11), Llama 4 Maverick: 31.6 (#105)

BenchmarkGPT-5.6 TerraLlama 4 Maverick
LMArena Vision12711142
GeoBench—52%
Blueprint-Bench 230.8%—
Furniture Assembly54.2%—
LMArena Document1472—
SpatialViz-Bench—31.8%

Multilingual GPT-5.6 Terra leads

GPT-5.6 Terra: 54.4 (#44), Llama 4 Maverick: 42.2 (#195)

BenchmarkGPT-5.6 TerraLlama 4 Maverick
LMArena Non-English14391269
LMArena Chinese15131277
LMArena French14711259
LMArena German14601291
LMArena Japanese14571207
LMArena Korean14251203
LMArena Russian14501286
LMArena Spanish14481293

Instruction Following GPT-5.6 Terra leads

GPT-5.6 Terra: 76.4 (#40), Llama 4 Maverick: 71.7 (#146)

BenchmarkGPT-5.6 TerraLlama 4 Maverick
LMArena Instruction Following14541267
IFEval—90.8%

Long Context GPT-5.6 Terra leads

GPT-5.6 Terra: 44.4 (#68), Llama 4 Maverick: 31.4 (#279)

BenchmarkGPT-5.6 TerraLlama 4 Maverick
LMArena Longer Query14511280
Fiction.LiveBench—46.2%

Writing & Preference GPT-5.6 Terra leads

GPT-5.6 Terra: 70.2 (#23), Llama 4 Maverick: 38.8 (#252)

BenchmarkGPT-5.6 TerraLlama 4 Maverick
LMArena Text14471287
LMArena Creative Writing14101267
EQ-Bench Creative Writing1855860
LMArena Multi-Turn14491289
Short-Story Creative Writing—62%
WildBench—80%
EQ-Bench 41234—

Frequently asked questions

Is GPT-5.6 Terra better than Llama 4 Maverick?

GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 30.9 on the Noometry Index. Llama 4 Maverick costs 15× less per token, which makes it the better buy when GPT-5.6 Terra's lead doesn't matter for your workload.

Which is cheaper, GPT-5.6 Terra 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 Terra lists at $2 and $12.

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

GPT-5.6 Terra scores higher on coding benchmarks: 57.7 versus 26.6 in the Noometry coding category.

Which has the bigger context window?

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

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

33 benchmarks have published results for both models. GPT-5.6 Terra has 52 scored results on Noometry and Llama 4 Maverick has 54.

Related comparisons

Go deeper