Model comparison
Granite 4.2 30b vs Qwen3 32B
Granite 4.2 30b is the stronger model overall, scoring 41.4 to 38.8 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 5 categories and Qwen3 32B in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Granite 4.2 30b leads 26.7 to 19.6.
Side by side
| Granite 4.2 30b | Qwen3 32B | |
|---|---|---|
| Provider | IBM | Alibaba (Qwen) |
| Noometry Index | 41.4 | 38.8 |
| Released | — | 2025-04 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.70 |
| Output $ / M tokens | — | $2.80 |
| Results tracked | 11 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Granite 4.2 30b leads
Granite 4.2 30b: 40.4 (#127), Qwen3 32B: 37.3 (#184)
| Benchmark | Granite 4.2 30b | Qwen3 32B |
|---|---|---|
| LMArena Coding | 1394 | 1358 |
| Aider Polyglot | — | 40% |
| SciCode | — | 35.4% |
Agentic & Tool Use Not comparable
Granite 4.2 30b: —, Qwen3 32B: 32.3 (#58)
| Benchmark | Granite 4.2 30b | Qwen3 32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 48.7% |
Reasoning Granite 4.2 30b leads
Granite 4.2 30b: 26.7 (#114), Qwen3 32B: 19.6 (#233)
| Benchmark | Granite 4.2 30b | Qwen3 32B |
|---|---|---|
| LMArena Hard Prompts | 1375 | 1335 |
| Kagi LLM Benchmark | — | 54.9% |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 5% |
| DTBench | — | 67.5% |
| LMCA | — | 17.3% |
| Epoch Capabilities Index | — | 138.51 |
Math Not comparable
Granite 4.2 30b: —, Qwen3 32B: 37.9 (#89)
| Benchmark | Granite 4.2 30b | Qwen3 32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 66.9% |
| LMArena Math | — | 1399 |
Knowledge Qwen3 32B leads
Granite 4.2 30b: 38.5 (#134), Qwen3 32B: 40.0 (#112)
| Benchmark | Granite 4.2 30b | Qwen3 32B |
|---|---|---|
| LMArena Expert | 1405 | 1362 |
| GPQA Diamond | — | 65.7% |
| Vectara Hallucination Rate | — | 5.9% |
Multilingual Granite 4.2 30b leads
Granite 4.2 30b: 47.4 (#146), Qwen3 32B: 45.8 (#162)
| Benchmark | Granite 4.2 30b | Qwen3 32B |
|---|---|---|
| LMArena Non-English | 1340 | 1317 |
| LMArena Chinese | 1415 | 1357 |
| LMArena Russian | 1342 | 1311 |
| LMArena German | — | 1341 |
Instruction Following Granite 4.2 30b leads
Granite 4.2 30b: 71.3 (#150), Qwen3 32B: 69.1 (#174)
| Benchmark | Granite 4.2 30b | Qwen3 32B |
|---|---|---|
| LMArena Instruction Following | 1346 | 1305 |
Long Context Qwen3 32B leads
Granite 4.2 30b: 41.4 (#135), Qwen3 32B: 43.7 (#83)
| Benchmark | Granite 4.2 30b | Qwen3 32B |
|---|---|---|
| LMArena Longer Query | 1359 | 1327 |
| Fiction.LiveBench | — | 74.2% |
Writing & Preference Granite 4.2 30b leads
Granite 4.2 30b: 54.3 (#147), Qwen3 32B: 53.1 (#158)
| Benchmark | Granite 4.2 30b | Qwen3 32B |
|---|---|---|
| LMArena Text | 1363 | 1340 |
| LMArena Creative Writing | 1294 | 1297 |
| LMArena Multi-Turn | 1340 | 1331 |
Frequently asked questions
Is Granite 4.2 30b better than Qwen3 32B?
Granite 4.2 30b is the stronger model overall, scoring 41.4 to 38.8 on the Noometry Index.
Is Granite 4.2 30b or Qwen3 32B better for coding?
Granite 4.2 30b scores higher on coding benchmarks: 40.4 versus 37.3 in the Noometry coding category.
How many benchmarks do Granite 4.2 30b and Qwen3 32B share?
11 benchmarks have published results for both models. Granite 4.2 30b has 11 scored results on Noometry and Qwen3 32B has 26.