Model comparison
Qwen2.5-Coder-32B vs Qwen3.5 35B-A3B
Qwen3.5 35B-A3B is the stronger model overall, scoring 42.0 to 33.4 on the Noometry Index.
Last verified . 13 shared benchmarks.
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. Qwen2.5-Coder-32B scores higher in 0 categories and Qwen3.5 35B-A3B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen3.5 35B-A3B leads 57.9 to 41.6.
- Qwen3.5 35B-A3B is cheaper at $0.25 / $2 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- Qwen3.5 35B-A3B accepts more context: 262K tokens versus 33K.
Side by side
| Qwen2.5-Coder-32B | Qwen3.5 35B-A3B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 33.4 | 42.0 |
| Released | 2024-09-18 | 2026-02-01 |
| Weights | Open | Open |
| Context window | 33K | 262K |
| Max output | 29K | 66K |
| Input $ / M tokens | $0.66 | $0.25 |
| Output $ / M tokens | $1 | $2 |
| Results tracked | 31 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5 35B-A3B leads
Qwen2.5-Coder-32B: 22.6 (#333), Qwen3.5 35B-A3B: 33.8 (#251)
| Benchmark | Qwen2.5-Coder-32B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Coding | 1276 | 1410 |
| SWE-bench Verified (bash only) | 9% | — |
| Aider Polyglot | 16.4% | — |
| LMArena WebDev | — | 1254 |
| SciCode | — | 29.3% |
| BigCodeBench Instruct | 49% | — |
| LiveBench Coding | 56.9% | — |
| BigCodeBench Complete | 58% | — |
| HumanEval+ | 87.2% | — |
| MBPP+ | 77% | — |
Reasoning Qwen3.5 35B-A3B leads
Qwen2.5-Coder-32B: 21.2 (#225), Qwen3.5 35B-A3B: 24.6 (#161)
| Benchmark | Qwen2.5-Coder-32B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1400 |
| Epoch Capabilities Index | 119.49 | 142.52 |
| CritPt | — | 0.6% |
| Chess Puzzles | — | 10% |
| LiveBench Reasoning | 42.1% | — |
| DTBench | — | 80% |
| LiveBench Data Analysis | 49.9% | — |
| LMCA | — | 29.5% |
| HellaSwag | 83% | — |
| LiveBench | 46.2% | — |
| WinoGrande | 80.8% | — |
Math Qwen3.5 35B-A3B leads
Qwen2.5-Coder-32B: 33.3 (#204), Qwen3.5 35B-A3B: 39.9 (#97)
| Benchmark | Qwen2.5-Coder-32B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Math | 1251 | 1404 |
| MathArena Final-Answer Competitions | — | 56% |
| OTIS Mock AIME 2024-2025 | — | 70% |
| LiveBench Math | 46.6% | — |
| GSM8K | 93% | — |
Knowledge Qwen3.5 35B-A3B leads
Qwen2.5-Coder-32B: 33.4 (#203), Qwen3.5 35B-A3B: 47.8 (#79)
| Benchmark | Qwen2.5-Coder-32B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Expert | 1221 | 1408 |
| GPQA Diamond | — | 83.5% |
| Vectara Hallucination Rate | — | 10.5% |
| ARC (AI2) Challenge | 70.5% | — |
| MMLU | 79.1% | — |
Multilingual Qwen3.5 35B-A3B leads
Qwen2.5-Coder-32B: 37.8 (#235), Qwen3.5 35B-A3B: 50.0 (#127)
| Benchmark | Qwen2.5-Coder-32B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Non-English | 1205 | 1378 |
| LMArena Chinese | 1222 | 1457 |
| LMArena Russian | 1228 | 1376 |
| LMArena French | — | 1412 |
| LMArena German | — | 1367 |
| LMArena Japanese | — | 1325 |
| LMArena Korean | — | 1356 |
| LMArena Spanish | — | 1392 |
Instruction Following Qwen3.5 35B-A3B leads
Qwen2.5-Coder-32B: 61.4 (#245), Qwen3.5 35B-A3B: 72.8 (#128)
| Benchmark | Qwen2.5-Coder-32B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Instruction Following | 1223 | 1379 |
| LiveBench Instruction Following | 58.7% | — |
Long Context Qwen3.5 35B-A3B leads
Qwen2.5-Coder-32B: 38.0 (#208), Qwen3.5 35B-A3B: 42.4 (#127)
| Benchmark | Qwen2.5-Coder-32B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Longer Query | 1251 | 1389 |
Writing & Preference Qwen3.5 35B-A3B leads
Qwen2.5-Coder-32B: 41.6 (#240), Qwen3.5 35B-A3B: 57.9 (#124)
| Benchmark | Qwen2.5-Coder-32B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Text | 1230 | 1395 |
| LMArena Creative Writing | 1174 | 1346 |
| LMArena Multi-Turn | 1222 | 1390 |
| LiveBench Language | 23.3% | — |
Frequently asked questions
Is Qwen2.5-Coder-32B better than Qwen3.5 35B-A3B?
Qwen3.5 35B-A3B is the stronger model overall, scoring 42.0 to 33.4 on the Noometry Index.
Which is cheaper, Qwen2.5-Coder-32B or Qwen3.5 35B-A3B?
Qwen3.5 35B-A3B is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is Qwen2.5-Coder-32B or Qwen3.5 35B-A3B better for coding?
Qwen3.5 35B-A3B scores higher on coding benchmarks: 33.8 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
Qwen3.5 35B-A3B does, with 262K tokens against 33K.
How many benchmarks do Qwen2.5-Coder-32B and Qwen3.5 35B-A3B share?
13 benchmarks have published results for both models. Qwen2.5-Coder-32B has 31 scored results on Noometry and Qwen3.5 35B-A3B has 28.