Model comparison
Qwen2.5-Max vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 40.7 on the Noometry Index.
Last verified . 18 shared benchmarks.
Alibaba (Qwen)Summary
- They share 18 benchmarks with published results for both. Qwen2.5-Max scores higher in 0 categories and Qwen3.8 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.2 to 36.9.
Side by side
| Qwen2.5-Max | Qwen3.8 Max | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 40.7 | 56.8 |
| Released | 2025-01-25 | 2026-08-02 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1M |
| Max output | — | 131K |
| Input $ / M tokens | — | $2 |
| Output $ / M tokens | — | $6 |
| Results tracked | 27 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
Qwen2.5-Max: 41.8 (#117), Qwen3.8 Max: 53.5 (#29)
| Benchmark | Qwen2.5-Max | Qwen3.8 Max |
|---|---|---|
| LMArena Coding | 1359 | 1502 |
| DeepSWE | — | 57.5% |
| LMArena WebDev | — | 1674 |
| FrontierSWE | — | 17.8% |
| SciCode | — | 53.2% |
| LiveBench Coding | 64.4% | — |
Agentic & Tool Use Not comparable
Qwen2.5-Max: —, Qwen3.8 Max: 45.4 (#14)
| Benchmark | Qwen2.5-Max | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | — | 63.3% |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
Reasoning Qwen3.8 Max leads
Qwen2.5-Max: 25.6 (#147), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Qwen2.5-Max | Qwen3.8 Max |
|---|---|---|
| LMArena Hard Prompts | 1360 | 1496 |
| Epoch Capabilities Index | 132.53 | 156.41 |
| NYT Connections (extended) | — | 88.3% |
| CritPt | — | 20% |
| Chess Puzzles | — | 40% |
| LiveBench Reasoning | 51.4% | — |
| Mystery Game Puzzles | — | 38% |
| DTBench | — | 92% |
| LiveBench Data Analysis | 67.9% | — |
| LMCA | — | 46.2% |
| LiveBench | 62.3% | — |
Math Qwen3.8 Max leads
Qwen2.5-Max: 36.9 (#162), Qwen3.8 Max: 73.2 (#20)
| Benchmark | Qwen2.5-Max | Qwen3.8 Max |
|---|---|---|
| LMArena Math | 1369 | 1499 |
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| OTIS Mock AIME 2024-2025 | — | 100% |
| ProofBench | — | 58% |
| LiveBench Math | 58.4% | — |
Knowledge Qwen3.8 Max leads
Qwen2.5-Max: 35.3 (#186), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Qwen2.5-Max | Qwen3.8 Max |
|---|---|---|
| LMArena Expert | 1337 | 1507 |
| GPQA Diamond | — | 92.7% |
| SimpleQA Verified | — | 47.3% |
| Confabulations | 21.8% | — |
Multimodal Not comparable
Qwen2.5-Max: —, Qwen3.8 Max: 37.2 (#75)
| Benchmark | Qwen2.5-Max | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Qwen3.8 Max leads
Qwen2.5-Max: 48.1 (#146), Qwen3.8 Max: 56.7 (#18)
| Benchmark | Qwen2.5-Max | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1352 | 1472 |
| LMArena Chinese | 1382 | 1538 |
| LMArena French | 1396 | 1503 |
| LMArena German | 1350 | 1483 |
| LMArena Japanese | 1300 | 1467 |
| LMArena Korean | 1304 | 1461 |
| LMArena Russian | 1353 | 1481 |
| LMArena Spanish | 1377 | 1492 |
Instruction Following Qwen3.8 Max leads
Qwen2.5-Max: 71.3 (#152), Qwen3.8 Max: 77.6 (#17)
| Benchmark | Qwen2.5-Max | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1335 | 1479 |
| LiveBench Instruction Following | 75.3% | — |
Long Context Qwen3.8 Max leads
Qwen2.5-Max: 41.4 (#142), Qwen3.8 Max: 45.6 (#31)
| Benchmark | Qwen2.5-Max | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1358 | 1489 |
Writing & Preference Qwen3.8 Max leads
Qwen2.5-Max: 55.4 (#146), Qwen3.8 Max: 67.1 (#30)
| Benchmark | Qwen2.5-Max | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1367 | 1483 |
| LMArena Creative Writing | 1339 | 1479 |
| LMArena Multi-Turn | 1364 | 1489 |
| Short-Story Creative Writing | 72.9% | — |
| LiveBench Language | 56.3% | — |
Frequently asked questions
Is Qwen2.5-Max better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 40.7 on the Noometry Index.
Is Qwen2.5-Max or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 41.8 in the Noometry coding category.
How many benchmarks do Qwen2.5-Max and Qwen3.8 Max share?
18 benchmarks have published results for both models. Qwen2.5-Max has 27 scored results on Noometry and Qwen3.8 Max has 39.