Model comparison
Granite 4.2 30b vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 41.8 on the Noometry Index.
Last verified . 11 shared benchmarks.
Alibaba (Qwen)Summary
- They share 11 benchmarks with published results for both. Granite 4.2 30b scores higher in 0 categories and Qwen3.8 Max in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 Max leads 54.4 to 27.8.
- Granite 4.2 30b has downloadable open weights; the other is API-only.
Side by side
| Granite 4.2 30b | Qwen3.8 Max | |
|---|---|---|
| Provider | IBM | Alibaba (Qwen) |
| Noometry Index | 41.8 | 56.8 |
| Released | — | 2026-08-02 |
| Weights | Open | Proprietary |
| Context window | — | 1M |
| Max output | — | 131K |
| Input $ / M tokens | — | $2 |
| Output $ / M tokens | — | $6 |
| Results tracked | 11 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
Granite 4.2 30b: 41.0 (#126), Qwen3.8 Max: 53.5 (#29)
| Benchmark | Granite 4.2 30b | Qwen3.8 Max |
|---|---|---|
| LMArena Coding | 1396 | 1502 |
| DeepSWE | — | 57.5% |
| LMArena WebDev | — | 1674 |
| FrontierSWE | — | 17.8% |
| SciCode | — | 53.2% |
Agentic & Tool Use Not comparable
Granite 4.2 30b: —, Qwen3.8 Max: 45.4 (#14)
| Benchmark | Granite 4.2 30b | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | — | 63.3% |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
Reasoning Qwen3.8 Max leads
Granite 4.2 30b: 27.8 (#112), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Granite 4.2 30b | Qwen3.8 Max |
|---|---|---|
| LMArena Hard Prompts | 1374 | 1496 |
| NYT Connections (extended) | — | 88.3% |
| CritPt | — | 20% |
| Chess Puzzles | — | 40% |
| Mystery Game Puzzles | — | 38% |
| DTBench | — | 92% |
| LMCA | — | 46.2% |
| Epoch Capabilities Index | — | 156.41 |
Math Not comparable
Granite 4.2 30b: —, Qwen3.8 Max: 73.2 (#20)
| Benchmark | Granite 4.2 30b | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| OTIS Mock AIME 2024-2025 | — | 100% |
| ProofBench | — | 58% |
| LMArena Math | — | 1499 |
Knowledge Qwen3.8 Max leads
Granite 4.2 30b: 39.1 (#138), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Granite 4.2 30b | Qwen3.8 Max |
|---|---|---|
| LMArena Expert | 1406 | 1507 |
| GPQA Diamond | — | 92.7% |
| SimpleQA Verified | — | 47.3% |
Multimodal Not comparable
Granite 4.2 30b: —, Qwen3.8 Max: 37.2 (#75)
| Benchmark | Granite 4.2 30b | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Qwen3.8 Max leads
Granite 4.2 30b: 47.3 (#151), Qwen3.8 Max: 56.7 (#18)
| Benchmark | Granite 4.2 30b | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1340 | 1472 |
| LMArena Chinese | 1414 | 1538 |
| LMArena Russian | 1343 | 1481 |
| LMArena French | — | 1503 |
| LMArena German | — | 1483 |
| LMArena Japanese | — | 1467 |
| LMArena Korean | — | 1461 |
| LMArena Spanish | — | 1492 |
Instruction Following Qwen3.8 Max leads
Granite 4.2 30b: 71.2 (#155), Qwen3.8 Max: 77.6 (#17)
| Benchmark | Granite 4.2 30b | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1347 | 1479 |
Long Context Qwen3.8 Max leads
Granite 4.2 30b: 41.4 (#140), Qwen3.8 Max: 45.6 (#31)
| Benchmark | Granite 4.2 30b | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1359 | 1489 |
Writing & Preference Qwen3.8 Max leads
Granite 4.2 30b: 53.8 (#156), Qwen3.8 Max: 67.1 (#30)
| Benchmark | Granite 4.2 30b | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1361 | 1483 |
| LMArena Creative Writing | 1288 | 1479 |
| LMArena Multi-Turn | 1339 | 1489 |
Frequently asked questions
Is Granite 4.2 30b better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 41.8 on the Noometry Index.
Is Granite 4.2 30b or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 41.0 in the Noometry coding category.
How many benchmarks do Granite 4.2 30b and Qwen3.8 Max share?
11 benchmarks have published results for both models. Granite 4.2 30b has 11 scored results on Noometry and Qwen3.8 Max has 39.