Model comparison
Llama 3.2 3B vs Qwen2.5-Max
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 28.9 on the Noometry Index.
Last verified . 13 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. Llama 3.2 3B scores higher in 0 categories and Qwen2.5-Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen2.5-Max leads 55.4 to 24.7.
- Llama 3.2 3B has downloadable open weights; the other is API-only.
Side by side
| Llama 3.2 3B | Qwen2.5-Max | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 28.9 | 40.7 |
| Released | 2024-09-24 | 2025-01-25 |
| Weights | Open | Proprietary |
| Context window | 131K | — |
| Max output | 118K | — |
| Input $ / M tokens | $0.05 | — |
| Output $ / M tokens | $0.33 | — |
| Results tracked | 18 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5-Max leads
Llama 3.2 3B: 27.6 (#319), Qwen2.5-Max: 41.8 (#117)
| Benchmark | Llama 3.2 3B | Qwen2.5-Max |
|---|---|---|
| LMArena Coding | 1098 | 1359 |
| BigCodeBench Instruct | 23.4% | — |
| LiveBench Coding | — | 64.4% |
| BigCodeBench Complete | 28.3% | — |
Agentic & Tool Use Not comparable
Llama 3.2 3B: 20.1 (#143), Qwen2.5-Max: —
| Benchmark | Llama 3.2 3B | Qwen2.5-Max |
|---|---|---|
| Berkeley Function Calling Leaderboard | 21.9% | — |
| BALROG | 10.1% | — |
Reasoning Qwen2.5-Max leads
Llama 3.2 3B: 21.0 (#228), Qwen2.5-Max: 25.6 (#147)
| Benchmark | Llama 3.2 3B | Qwen2.5-Max |
|---|---|---|
| LMArena Hard Prompts | 1095 | 1360 |
| LiveBench Reasoning | — | 51.4% |
| LiveBench Data Analysis | — | 67.9% |
| Epoch Capabilities Index | — | 132.53 |
| LiveBench | — | 62.3% |
Math Qwen2.5-Max leads
Llama 3.2 3B: 32.4 (#214), Qwen2.5-Max: 36.9 (#162)
| Benchmark | Llama 3.2 3B | Qwen2.5-Max |
|---|---|---|
| LMArena Math | 1126 | 1369 |
| LiveBench Math | — | 58.4% |
Knowledge Qwen2.5-Max leads
Llama 3.2 3B: 29.7 (#235), Qwen2.5-Max: 35.3 (#186)
| Benchmark | Llama 3.2 3B | Qwen2.5-Max |
|---|---|---|
| LMArena Expert | 1090 | 1337 |
| Confabulations | — | 21.8% |
Multilingual Qwen2.5-Max leads
Llama 3.2 3B: 26.2 (#281), Qwen2.5-Max: 48.1 (#146)
| Benchmark | Llama 3.2 3B | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1019 | 1352 |
| LMArena Chinese | 1017 | 1382 |
| LMArena German | 1056 | 1350 |
| LMArena Russian | 949 | 1353 |
| LMArena French | — | 1396 |
| LMArena Japanese | — | 1300 |
| LMArena Korean | — | 1304 |
| LMArena Spanish | — | 1377 |
Instruction Following Qwen2.5-Max leads
Llama 3.2 3B: 56.0 (#275), Qwen2.5-Max: 71.3 (#152)
| Benchmark | Llama 3.2 3B | Qwen2.5-Max |
|---|---|---|
| LMArena Instruction Following | 1089 | 1335 |
| LiveBench Instruction Following | — | 75.3% |
Long Context Qwen2.5-Max leads
Llama 3.2 3B: 33.4 (#261), Qwen2.5-Max: 41.4 (#142)
| Benchmark | Llama 3.2 3B | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1100 | 1358 |
Writing & Preference Qwen2.5-Max leads
Llama 3.2 3B: 24.7 (#307), Qwen2.5-Max: 55.4 (#146)
| Benchmark | Llama 3.2 3B | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1110 | 1367 |
| LMArena Creative Writing | 1094 | 1339 |
| LMArena Multi-Turn | 1105 | 1364 |
| Short-Story Creative Writing | — | 72.9% |
| EQ-Bench Creative Writing | 595 | — |
| LiveBench Language | — | 56.3% |
Frequently asked questions
Is Llama 3.2 3B better than Qwen2.5-Max?
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 28.9 on the Noometry Index.
Is Llama 3.2 3B or Qwen2.5-Max better for coding?
Qwen2.5-Max scores higher on coding benchmarks: 41.8 versus 27.6 in the Noometry coding category.
How many benchmarks do Llama 3.2 3B and Qwen2.5-Max share?
13 benchmarks have published results for both models. Llama 3.2 3B has 18 scored results on Noometry and Qwen2.5-Max has 27.