Model comparison

Llama 3.2 90B vs Qwen Max

Qwen Max is the stronger model overall, scoring 34.7 to 27.5 on the Noometry Index.

Last verified October 10, 2026. 3 shared benchmarks.

Summary

They share 3 benchmarks with published results for both. Llama 3.2 90B scores higher in 0 categories and Qwen Max in 3 categories; 3 gaps are clear of the uncertainty.
The widest gap is in math, where Qwen Max leads 22.3 to 11.1.
The biggest single-benchmark swing is MATH Level 5: 39.4% for Llama 3.2 90B and 67.2% for Qwen Max.
Llama 3.2 90B has downloadable open weights; the other is API-only.

Side by side

Llama 3.2 90B and Qwen Max specifications
	Llama 3.2 90B	Qwen Max
Provider	Meta	Alibaba (Qwen)
Noometry Index	27.5	34.7
Released	2024-09-24	2024-04-03
Weights	Open	Proprietary
Context window	—	33K
Max output	—	8K
Input $ / M tokens	—	$1.60
Output $ / M tokens	—	$6.40
Results tracked	9	23

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

Category by category

Coding Not comparable

Llama 3.2 90B: —, Qwen Max: 30.7 (#292)

Coding benchmarks
Benchmark	Llama 3.2 90B	Qwen Max
Aider Polyglot	—	21.8%
LMArena Coding	—	1288

Agentic & Tool Use Not comparable

Llama 3.2 90B: 30.0 (#80), Qwen Max: —

Agentic & Tool Use benchmarks
Benchmark	Llama 3.2 90B	Qwen Max
BALROG	27.3%	—

Reasoning Qwen Max leads

Llama 3.2 90B: 21.7 (#217), Qwen Max: 25.1 (#151)

Reasoning benchmarks
Benchmark	Llama 3.2 90B	Qwen Max
EnigmaEval	0.4%	—
LMArena Hard Prompts	—	1269
Epoch Capabilities Index	125.5	—

Math Qwen Max leads

Llama 3.2 90B: 11.1 (#308), Qwen Max: 22.3 (#276)

Math benchmarks
Benchmark	Llama 3.2 90B	Qwen Max
OTIS Mock AIME 2024-2025	2.6%	16.1%
MATH Level 5	39.4%	67.2%
LMArena Math	—	1275
FrontierMath (Feb 2025 set)	—	1%

Knowledge Qwen Max leads

Llama 3.2 90B: 21.7 (#274), Qwen Max: 30.3 (#228)

Knowledge benchmarks
Benchmark	Llama 3.2 90B	Qwen Max
GPQA Diamond	41%	56.1%
LMArena Expert	—	1248
MMLU	80.3%	—

Multimodal Not comparable

Llama 3.2 90B: 25.4 (#124), Qwen Max: —

Multimodal benchmarks
Benchmark	Llama 3.2 90B	Qwen Max
LMArena Vision	1000	—
GeoBench	52%	—

Multilingual Not comparable

Llama 3.2 90B: —, Qwen Max: 41.8 (#202)

Multilingual benchmarks
Benchmark	Llama 3.2 90B	Qwen Max
LMArena Non-English	—	1263
LMArena Chinese	—	1254
LMArena French	—	1330
LMArena German	—	1254
LMArena Japanese	—	1205
LMArena Korean	—	1142
LMArena Russian	—	1274
LMArena Spanish	—	1290

Instruction Following Not comparable

Llama 3.2 90B: —, Qwen Max: 66.5 (#208)

Instruction Following benchmarks
Benchmark	Llama 3.2 90B	Qwen Max
LMArena Instruction Following	—	1262

Long Context Not comparable

Llama 3.2 90B: —, Qwen Max: 39.4 (#180)

Long Context benchmarks
Benchmark	Llama 3.2 90B	Qwen Max
Fiction.LiveBench	—	66.7%
LMArena Longer Query	—	1288

Writing & Preference Not comparable

Llama 3.2 90B: —, Qwen Max: 47.8 (#205)

Writing & Preference benchmarks
Benchmark	Llama 3.2 90B	Qwen Max
LMArena Text	—	1282
LMArena Creative Writing	—	1248
LMArena Multi-Turn	—	1277

Frequently asked questions

Is Llama 3.2 90B better than Qwen Max?