Model comparison

Llama 3.2 90B vs Qwen3.8 Max

Qwen3.8 Max is the stronger model overall, scoring 56.8 to 27.5 on the Noometry Index.

Last verified October 10, 2026. 4 shared benchmarks.

Summary

They share 4 benchmarks with published results for both. Llama 3.2 90B scores higher in 0 categories and Qwen3.8 Max in 5 categories; 5 gaps are clear of the uncertainty.
The widest gap is in math, where Qwen3.8 Max leads 73.2 to 11.1.
The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 2.6% for Llama 3.2 90B and 100% for Qwen3.8 Max.
Llama 3.2 90B has downloadable open weights; the other is API-only.

Side by side

Llama 3.2 90B and Qwen3.8 Max specifications
	Llama 3.2 90B	Qwen3.8 Max
Provider	Meta	Alibaba (Qwen)
Noometry Index	27.5	56.8
Released	2024-09-24	2026-08-02
Weights	Open	Proprietary
Context window	—	1M
Max output	—	131K
Input $ / M tokens	—	$2
Output $ / M tokens	—	$6
Results tracked	9	39

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

Category by category

Coding Not comparable

Llama 3.2 90B: —, Qwen3.8 Max: 53.5 (#29)

Coding benchmarks
Benchmark	Llama 3.2 90B	Qwen3.8 Max
DeepSWE	—	57.5%
LMArena WebDev	—	1674
FrontierSWE	—	17.8%
SciCode	—	53.2%
LMArena Coding	—	1502

Agentic & Tool Use Qwen3.8 Max leads

Llama 3.2 90B: 30.0 (#80), Qwen3.8 Max: 45.4 (#14)

Agentic & Tool Use benchmarks
Benchmark	Llama 3.2 90B	Qwen3.8 Max
APEX-Agents	—	63.3%
τ²-bench Banking	—	55.1%
BALROG	27.3%	—
GDP.pdf	—	23.2%

Reasoning Qwen3.8 Max leads

Llama 3.2 90B: 21.7 (#217), Qwen3.8 Max: 54.4 (#26)

Reasoning benchmarks
Benchmark	Llama 3.2 90B	Qwen3.8 Max
Epoch Capabilities Index	125.5	156.41
NYT Connections (extended)	—	88.3%
CritPt	—	20%
Chess Puzzles	—	40%
EnigmaEval	0.4%	—
LMArena Hard Prompts	—	1496
Mystery Game Puzzles	—	38%
DTBench	—	92%
LMCA	—	46.2%

Math Qwen3.8 Max leads

Llama 3.2 90B: 11.1 (#308), Qwen3.8 Max: 73.2 (#20)

Math benchmarks
Benchmark	Llama 3.2 90B	Qwen3.8 Max
OTIS Mock AIME 2024-2025	2.6%	100%
FrontierMath (Tiers 1-3)	—	74.7%
FrontierMath Tier 4	—	46.3%
ProofBench	—	58%
LMArena Math	—	1499
MATH Level 5	39.4%	—

Knowledge Qwen3.8 Max leads

Llama 3.2 90B: 21.7 (#274), Qwen3.8 Max: 61.7 (#27)

Knowledge benchmarks
Benchmark	Llama 3.2 90B	Qwen3.8 Max
GPQA Diamond	41%	92.7%
SimpleQA Verified	—	47.3%
LMArena Expert	—	1507
MMLU	80.3%	—

Multimodal Qwen3.8 Max leads

Llama 3.2 90B: 25.4 (#124), Qwen3.8 Max: 37.2 (#75)

Multimodal benchmarks
Benchmark	Llama 3.2 90B	Qwen3.8 Max
LMArena Vision	1000	1314
GeoBench	52%	—
Furniture Assembly	—	20%

Multilingual Not comparable

Llama 3.2 90B: —, Qwen3.8 Max: 56.7 (#18)

Multilingual benchmarks
Benchmark	Llama 3.2 90B	Qwen3.8 Max
LMArena Non-English	—	1472
LMArena Chinese	—	1538
LMArena French	—	1503
LMArena German	—	1483
LMArena Japanese	—	1467
LMArena Korean	—	1461
LMArena Russian	—	1481
LMArena Spanish	—	1492

Instruction Following Not comparable

Llama 3.2 90B: —, Qwen3.8 Max: 77.6 (#17)

Instruction Following benchmarks
Benchmark	Llama 3.2 90B	Qwen3.8 Max
LMArena Instruction Following	—	1479

Long Context Not comparable

Llama 3.2 90B: —, Qwen3.8 Max: 45.6 (#31)

Long Context benchmarks
Benchmark	Llama 3.2 90B	Qwen3.8 Max
LMArena Longer Query	—	1489

Writing & Preference Not comparable

Llama 3.2 90B: —, Qwen3.8 Max: 67.1 (#30)

Writing & Preference benchmarks
Benchmark	Llama 3.2 90B	Qwen3.8 Max
LMArena Text	—	1483
LMArena Creative Writing	—	1479
LMArena Multi-Turn	—	1489

Frequently asked questions

Is Llama 3.2 90B better than Qwen3.8 Max?