Model comparison
Llama 3.2 3B vs Qwen3.6 Max Preview
Qwen3.6 Max Preview is the stronger model overall, scoring 51.1 to 28.4 on the Noometry Index. Llama 3.2 3B costs 24× less per token, which makes it the better buy when Qwen3.6 Max Preview's lead doesn't matter for your workload.
Last verified . 12 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 12 benchmarks with published results for both. Llama 3.2 3B scores higher in 0 categories and Qwen3.6 Max Preview in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen3.6 Max Preview leads 64.0 to 24.9.
- Llama 3.2 3B is cheaper at $0.05 / $0.33 per million input/output tokens, against $1.30 / $7.80 for Qwen3.6 Max Preview.
- Qwen3.6 Max Preview accepts more context: 262K tokens versus 131K.
- Llama 3.2 3B has downloadable open weights; the other is API-only.
Side by side
| Llama 3.2 3B | Qwen3.6 Max Preview | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 28.4 | 51.1 |
| Released | 2024-09-24 | 2026-04-20 |
| Weights | Open | Proprietary |
| Context window | 131K | 262K |
| Max output | 118K | 66K |
| Input $ / M tokens | $0.05 | $1.30 |
| Output $ / M tokens | $0.33 | $7.80 |
| Results tracked | 18 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.6 Max Preview leads
Llama 3.2 3B: 27.1 (#314), Qwen3.6 Max Preview: 48.6 (#50)
| Benchmark | Llama 3.2 3B | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Coding | 1098 | 1472 |
| SWE-bench Verified | — | 76.7% |
| LMArena WebDev | — | 1482 |
| BigCodeBench Instruct | 23.4% | — |
| BigCodeBench Complete | 28.3% | — |
Agentic & Tool Use Not comparable
Llama 3.2 3B: 19.9 (#135), Qwen3.6 Max Preview: —
| Benchmark | Llama 3.2 3B | Qwen3.6 Max Preview |
|---|---|---|
| Berkeley Function Calling Leaderboard | 21.9% | — |
| BALROG | 10.1% | — |
| Vending-Bench 2 | — | 4,254 |
Reasoning Qwen3.6 Max Preview leads
Llama 3.2 3B: 20.1 (#225), Qwen3.6 Max Preview: 40.8 (#47)
| Benchmark | Llama 3.2 3B | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Hard Prompts | 1095 | 1457 |
| SimpleBench | — | 63% |
| NYT Connections (extended) | — | 74.1% |
| Chess Puzzles | — | 20% |
| Mystery Game Puzzles | — | 19% |
| DTBench | — | 87.2% |
| LMCA | — | 42.5% |
| Epoch Capabilities Index | — | 149.27 |
Math Qwen3.6 Max Preview leads
Llama 3.2 3B: 29.9 (#213), Qwen3.6 Max Preview: 52.2 (#40)
| Benchmark | Llama 3.2 3B | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Math | 1126 | 1465 |
| OTIS Mock AIME 2024-2025 | — | 91.1% |
| FrontierMath (Feb 2025 set) | — | 23.1% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Qwen3.6 Max Preview leads
Llama 3.2 3B: 29.1 (#232), Qwen3.6 Max Preview: 57.6 (#35)
| Benchmark | Llama 3.2 3B | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Expert | 1090 | 1476 |
| GPQA Diamond | — | 87.4% |
| SimpleQA Verified | — | 52% |
Multilingual Qwen3.6 Max Preview leads
Llama 3.2 3B: 26.4 (#276), Qwen3.6 Max Preview: 54.3 (#46)
| Benchmark | Llama 3.2 3B | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Non-English | 1019 | 1436 |
| LMArena Chinese | 1017 | 1487 |
| LMArena Russian | 949 | 1444 |
| LMArena French | — | 1449 |
| LMArena German | 1056 | — |
| LMArena Spanish | — | 1452 |
Instruction Following Qwen3.6 Max Preview leads
Llama 3.2 3B: 56.3 (#270), Qwen3.6 Max Preview: 75.8 (#51)
| Benchmark | Llama 3.2 3B | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Instruction Following | 1089 | 1437 |
Long Context Qwen3.6 Max Preview leads
Llama 3.2 3B: 33.4 (#256), Qwen3.6 Max Preview: 44.6 (#59)
| Benchmark | Llama 3.2 3B | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Longer Query | 1101 | 1457 |
Writing & Preference Qwen3.6 Max Preview leads
Llama 3.2 3B: 24.9 (#302), Qwen3.6 Max Preview: 64.0 (#58)
| Benchmark | Llama 3.2 3B | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Text | 1110 | 1447 |
| LMArena Creative Writing | 1094 | 1435 |
| LMArena Multi-Turn | 1105 | 1456 |
| EQ-Bench Creative Writing | 595 | — |
Frequently asked questions
Is Llama 3.2 3B better than Qwen3.6 Max Preview?
Qwen3.6 Max Preview is the stronger model overall, scoring 51.1 to 28.4 on the Noometry Index. Llama 3.2 3B costs 24× less per token, which makes it the better buy when Qwen3.6 Max Preview's lead doesn't matter for your workload.
Which is cheaper, Llama 3.2 3B or Qwen3.6 Max Preview?
Llama 3.2 3B is cheaper. It lists at $0.05 per million input tokens and $0.33 per million output tokens; Qwen3.6 Max Preview lists at $1.30 and $7.80.
Is Llama 3.2 3B or Qwen3.6 Max Preview better for coding?
Qwen3.6 Max Preview scores higher on coding benchmarks: 48.6 versus 27.1 in the Noometry coding category.
Which has the bigger context window?
Qwen3.6 Max Preview does, with 262K tokens against 131K.
How many benchmarks do Llama 3.2 3B and Qwen3.6 Max Preview share?
12 benchmarks have published results for both models. Llama 3.2 3B has 18 scored results on Noometry and Qwen3.6 Max Preview has 29.