Model comparison
Qwen2-72B vs Qwen3.6 35B-A3B
Qwen3.6 35B-A3B is the stronger model overall, scoring 37.6 to 30.0 on the Noometry Index.
Last verified . 3 shared benchmarks.
Alibaba (Qwen)Summary
- They share 3 benchmarks with published results for both. Qwen2-72B scores higher in 0 categories and Qwen3.6 35B-A3B in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.6 35B-A3B leads 51.3 to 21.2.
- The biggest single-benchmark swing is GPQA Diamond: 40.8% for Qwen2-72B and 84.8% for Qwen3.6 35B-A3B.
Side by side
| Qwen2-72B | Qwen3.6 35B-A3B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 30.0 | 37.6 |
| Released | 2024-06-07 | 2026-04-01 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $1.49 |
| Results tracked | 26 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.6 35B-A3B leads
Qwen2-72B: 29.1 (#310), Qwen3.6 35B-A3B: 37.2 (#196)
| Benchmark | Qwen2-72B | Qwen3.6 35B-A3B |
|---|---|---|
| WeirdML | 11.3% | 34.5% |
| SciCode | — | 35.8% |
| BigCodeBench Instruct | 38.5% | — |
| LMArena Coding | 1196 | — |
| BigCodeBench Complete | 54% | — |
Agentic & Tool Use Qwen3.6 35B-A3B leads
Qwen2-72B: 17.0 (#146), Qwen3.6 35B-A3B: 22.1 (#134)
| Benchmark | Qwen2-72B | Qwen3.6 35B-A3B |
|---|---|---|
| Terminal-Bench | — | 23% |
| TheAgentCompany | 1.1% | — |
| METR Time Horizons | 29.9% | — |
Reasoning Qwen3.6 35B-A3B leads
Qwen2-72B: 23.2 (#181), Qwen3.6 35B-A3B: 28.0 (#109)
| Benchmark | Qwen2-72B | Qwen3.6 35B-A3B |
|---|---|---|
| Epoch Capabilities Index | 125.28 | 143.93 |
| NYT Connections (extended) | — | 41.6% |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 26% |
| LMArena Hard Prompts | 1191 | — |
| Mystery Game Puzzles | — | 22% |
| DTBench | — | 73.9% |
| LMCA | — | 29.7% |
| Surface Evolver Bench | — | 44.4% |
Math Qwen3.6 35B-A3B leads
Qwen2-72B: 30.2 (#236), Qwen3.6 35B-A3B: 38.9 (#121)
| Benchmark | Qwen2-72B | Qwen3.6 35B-A3B |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 20.4% |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| LMArena Math | 1235 | — |
| MATH Level 5 | 39.1% | — |
Knowledge Qwen3.6 35B-A3B leads
Qwen2-72B: 21.2 (#275), Qwen3.6 35B-A3B: 51.3 (#68)
| Benchmark | Qwen2-72B | Qwen3.6 35B-A3B |
|---|---|---|
| GPQA Diamond | 40.8% | 84.8% |
| LMArena Expert | 1171 | — |
| MMLU | 82.4% | — |
Multilingual Not comparable
Qwen2-72B: 35.9 (#244), Qwen3.6 35B-A3B: —
| Benchmark | Qwen2-72B | Qwen3.6 35B-A3B |
|---|---|---|
| LMArena Non-English | 1176 | — |
| LMArena Chinese | 1240 | — |
| LMArena French | 1170 | — |
| LMArena German | 1151 | — |
| LMArena Japanese | 1111 | — |
| LMArena Korean | 1083 | — |
| LMArena Russian | 1169 | — |
| LMArena Spanish | 1169 | — |
Instruction Following Not comparable
Qwen2-72B: 61.7 (#241), Qwen3.6 35B-A3B: —
| Benchmark | Qwen2-72B | Qwen3.6 35B-A3B |
|---|---|---|
| LMArena Instruction Following | 1181 | — |
Long Context Not comparable
Qwen2-72B: 36.1 (#235), Qwen3.6 35B-A3B: —
| Benchmark | Qwen2-72B | Qwen3.6 35B-A3B |
|---|---|---|
| LMArena Longer Query | 1192 | — |
Writing & Preference Not comparable
Qwen2-72B: 40.8 (#241), Qwen3.6 35B-A3B: —
| Benchmark | Qwen2-72B | Qwen3.6 35B-A3B |
|---|---|---|
| LMArena Text | 1203 | — |
| LMArena Creative Writing | 1181 | — |
| LMArena Multi-Turn | 1196 | — |
Frequently asked questions
Is Qwen2-72B better than Qwen3.6 35B-A3B?
Qwen3.6 35B-A3B is the stronger model overall, scoring 37.6 to 30.0 on the Noometry Index.
Is Qwen2-72B or Qwen3.6 35B-A3B better for coding?
Qwen3.6 35B-A3B scores higher on coding benchmarks: 37.2 versus 29.1 in the Noometry coding category.
How many benchmarks do Qwen2-72B and Qwen3.6 35B-A3B share?
3 benchmarks have published results for both models. Qwen2-72B has 26 scored results on Noometry and Qwen3.6 35B-A3B has 14.