Model comparison

Llama 4 Maverick vs Qwen3-4B

Qwen3-4B is the stronger model overall, scoring 31.9 to 30.9 on the Noometry Index.

Last verified October 10, 2026. 4 shared benchmarks.

Summary

They share 4 benchmarks with published results for both. Llama 4 Maverick scores higher in 2 categories and Qwen3-4B in 2 categories; 2 gaps are clear of the uncertainty.
The widest gap is in reasoning, where Qwen3-4B leads 19.2 to 10.1.
The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 20.6% for Llama 4 Maverick and 52.2% for Qwen3-4B.

Side by side

Llama 4 Maverick and Qwen3-4B specifications
	Llama 4 Maverick	Qwen3-4B
Provider	Meta	Alibaba (Qwen)
Noometry Index	30.9	31.9
Released	2025-04-05	2025-04-29
Weights	Open	Open
Context window	128K	—
Max output	4K	—
Input $ / M tokens	$0.19	—
Output $ / M tokens	$0.65	—
Results tracked	54	6

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

Category by category

Coding Not comparable

Llama 4 Maverick: 26.6 (#324), Qwen3-4B: —

Coding benchmarks
Benchmark	Llama 4 Maverick	Qwen3-4B
SWE-bench Verified (bash only)	21%	—
Aider Polyglot	15.6%	—
SciCode	33.1%	—
WeirdML	24.5%	—
BigCodeBench Instruct	49.7%	—
LMArena Coding	1302	—
BigCodeBench Complete	61.4%	—
ALE-Bench	172.97	—

Agentic & Tool Use Too close to call

Llama 4 Maverick: 28.2 (#91), Qwen3-4B: 27.6 (#100)

Agentic & Tool Use benchmarks
Benchmark	Llama 4 Maverick	Qwen3-4B
Berkeley Function Calling Leaderboard	37.3%	35.7%

Reasoning Qwen3-4B leads

Llama 4 Maverick: 10.1 (#342), Qwen3-4B: 19.2 (#268)

Reasoning benchmarks
Benchmark	Llama 4 Maverick	Qwen3-4B
ARC-AGI-2	0%	—
SimpleBench	27.7%	—
Kagi LLM Benchmark	55.9%	—
NYT Connections (extended)	8%	—
ARC-AGI-1	4.4%	—
CritPt	0%	—
Chess Puzzles	—	4%
EnigmaEval	0.6%	—
LMArena Hard Prompts	1281	—
DTBench	61.9%	—
LMCA	15.9%	—
Epoch Capabilities Index	132.2	—
ForecastBench	57.5	—

Math Qwen3-4B leads

Llama 4 Maverick: 26.0 (#262), Qwen3-4B: 29.7 (#240)

Math benchmarks
Benchmark	Llama 4 Maverick	Qwen3-4B
OTIS Mock AIME 2024-2025	20.6%	52.2%
MathArena Final-Answer Competitions	—	38.5%
Omni-MATH	42.2%	—
LMArena Math	1299	—
MATH Level 5	73%	—
FrontierMath (Feb 2025 set)	0.7%	—

Knowledge Too close to call

Llama 4 Maverick: 33.4 (#204), Qwen3-4B: 33.0 (#208)

Knowledge benchmarks
Benchmark	Llama 4 Maverick	Qwen3-4B
GPQA Diamond	67%	52.3%
Vectara Hallucination Rate	8.2%	5.7%
Humanity's Last Exam	5.7%	—
MMLU-Pro	81%	—
Confabulations	22.6%	—
GPQA (HELM)	65%	—
LMArena Expert	1259	—

Multimodal Not comparable

Llama 4 Maverick: 31.6 (#105), Qwen3-4B: —

Multimodal benchmarks
Benchmark	Llama 4 Maverick	Qwen3-4B
LMArena Vision	1142	—
GeoBench	52%	—
SpatialViz-Bench	31.8%	—

Multilingual Not comparable

Llama 4 Maverick: 42.2 (#195), Qwen3-4B: —

Multilingual benchmarks
Benchmark	Llama 4 Maverick	Qwen3-4B
LMArena Non-English	1269	—
LMArena Chinese	1277	—
LMArena French	1259	—
LMArena German	1291	—
LMArena Japanese	1207	—
LMArena Korean	1203	—
LMArena Russian	1286	—
LMArena Spanish	1293	—

Instruction Following Not comparable

Llama 4 Maverick: 71.7 (#146), Qwen3-4B: —

Instruction Following benchmarks
Benchmark	Llama 4 Maverick	Qwen3-4B
IFEval	90.8%	—
LMArena Instruction Following	1267	—

Long Context Not comparable

Llama 4 Maverick: 31.4 (#279), Qwen3-4B: —

Long Context benchmarks
Benchmark	Llama 4 Maverick	Qwen3-4B
Fiction.LiveBench	46.2%	—
LMArena Longer Query	1280	—

Writing & Preference Not comparable

Llama 4 Maverick: 38.8 (#252), Qwen3-4B: —

Writing & Preference benchmarks
Benchmark	Llama 4 Maverick	Qwen3-4B
LMArena Text	1287	—
LMArena Creative Writing	1267	—
Short-Story Creative Writing	62%	—
EQ-Bench Creative Writing	860	—
WildBench	80%	—
LMArena Multi-Turn	1289	—

Frequently asked questions

Is Llama 4 Maverick better than Qwen3-4B?