Model comparison
Llama 3.2 90B vs Qwen3 32B
Qwen3 32B is the stronger model overall, scoring 39.2 to 27.5 on the Noometry Index.
Last verified . 3 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 3 benchmarks with published results for both. Llama 3.2 90B scores higher in 1 category and Qwen3 32B in 3 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3 32B leads 39.7 to 11.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 2.6% for Llama 3.2 90B and 66.9% for Qwen3 32B.
Side by side
| Llama 3.2 90B | Qwen3 32B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 27.5 | 39.2 |
| Released | 2024-09-24 | 2025-04 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.70 |
| Output $ / M tokens | — | $2.80 |
| Results tracked | 9 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Llama 3.2 90B: —, Qwen3 32B: 37.7 (#190)
| Benchmark | Llama 3.2 90B | Qwen3 32B |
|---|---|---|
| Aider Polyglot | — | 40% |
| SciCode | — | 35.4% |
| LMArena Coding | — | 1358 |
Agentic & Tool Use Qwen3 32B leads
Llama 3.2 90B: 30.0 (#80), Qwen3 32B: 32.6 (#62)
| Benchmark | Llama 3.2 90B | Qwen3 32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 48.7% |
| BALROG | 27.3% | — |
Reasoning Llama 3.2 90B leads
Llama 3.2 90B: 21.7 (#217), Qwen3 32B: 20.2 (#241)
| Benchmark | Llama 3.2 90B | Qwen3 32B |
|---|---|---|
| Epoch Capabilities Index | 125.5 | 138.51 |
| Kagi LLM Benchmark | — | 54.9% |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 5% |
| EnigmaEval | 0.4% | — |
| LMArena Hard Prompts | — | 1334 |
| DTBench | — | 67.5% |
| LMCA | — | 17.3% |
Math Qwen3 32B leads
Llama 3.2 90B: 11.1 (#308), Qwen3 32B: 39.7 (#99)
| Benchmark | Llama 3.2 90B | Qwen3 32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.6% | 66.9% |
| LMArena Math | — | 1399 |
| MATH Level 5 | 39.4% | — |
Knowledge Qwen3 32B leads
Llama 3.2 90B: 21.7 (#274), Qwen3 32B: 40.0 (#125)
| Benchmark | Llama 3.2 90B | Qwen3 32B |
|---|---|---|
| GPQA Diamond | 41% | 65.7% |
| Vectara Hallucination Rate | — | 5.9% |
| LMArena Expert | — | 1362 |
| MMLU | 80.3% | — |
Multimodal Not comparable
Llama 3.2 90B: 25.4 (#124), Qwen3 32B: —
| Benchmark | Llama 3.2 90B | Qwen3 32B |
|---|---|---|
| LMArena Vision | 1000 | — |
| GeoBench | 52% | — |
Multilingual Not comparable
Llama 3.2 90B: —, Qwen3 32B: 45.6 (#167)
| Benchmark | Llama 3.2 90B | Qwen3 32B |
|---|---|---|
| LMArena Non-English | — | 1317 |
| LMArena Chinese | — | 1357 |
| LMArena German | — | 1341 |
| LMArena Russian | — | 1311 |
Instruction Following Not comparable
Llama 3.2 90B: —, Qwen3 32B: 68.9 (#179)
| Benchmark | Llama 3.2 90B | Qwen3 32B |
|---|---|---|
| LMArena Instruction Following | — | 1305 |
Long Context Not comparable
Llama 3.2 90B: —, Qwen3 32B: 43.8 (#87)
| Benchmark | Llama 3.2 90B | Qwen3 32B |
|---|---|---|
| Fiction.LiveBench | — | 74.2% |
| LMArena Longer Query | — | 1327 |
Writing & Preference Not comparable
Llama 3.2 90B: —, Qwen3 32B: 52.9 (#163)
| Benchmark | Llama 3.2 90B | Qwen3 32B |
|---|---|---|
| LMArena Text | — | 1340 |
| LMArena Creative Writing | — | 1297 |
| LMArena Multi-Turn | — | 1331 |
Frequently asked questions
Is Llama 3.2 90B better than Qwen3 32B?
Qwen3 32B is the stronger model overall, scoring 39.2 to 27.5 on the Noometry Index.
How many benchmarks do Llama 3.2 90B and Qwen3 32B share?
3 benchmarks have published results for both models. Llama 3.2 90B has 9 scored results on Noometry and Qwen3 32B has 26.