Model comparison
Granite 4.0 Micro vs Qwen3-30B-A3B
Qwen3-30B-A3B is the stronger model overall, scoring 38.9 to 29.0 on the Noometry Index. Granite 4.0 Micro costs 5.3× less per token, which makes it the better buy when Qwen3-30B-A3B's lead doesn't matter for your workload.
Last verified . 3 shared benchmarks.
Alibaba (Qwen)Summary
- They share 3 benchmarks with published results for both. Granite 4.0 Micro scores higher in 0 categories and Qwen3-30B-A3B in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3-30B-A3B leads 41.8 to 9.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 2.8% for Granite 4.0 Micro and 70.3% for Qwen3-30B-A3B.
- Granite 4.0 Micro is cheaper at $0.017 / $0.11 per million input/output tokens, against $0.12 / $0.50 for Qwen3-30B-A3B.
- Granite 4.0 Micro accepts more context: 131K tokens versus 41K.
Side by side
| Granite 4.0 Micro | Qwen3-30B-A3B | |
|---|---|---|
| Provider | IBM | Alibaba (Qwen) |
| Noometry Index | 29.0 | 38.9 |
| Released | 2025-10-02 | 2025-04-28 |
| Weights | Open | Open |
| Context window | 131K | 41K |
| Max output | 118K | 16K |
| Input $ / M tokens | $0.017 | $0.12 |
| Output $ / M tokens | $0.11 | $0.50 |
| Results tracked | 8 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Granite 4.0 Micro: —, Qwen3-30B-A3B: 37.5 (#194)
| Benchmark | Granite 4.0 Micro | Qwen3-30B-A3B |
|---|---|---|
| SciCode | — | 33.3% |
| WeirdML | — | 29.8% |
| LMArena Coding | — | 1416 |
Agentic & Tool Use Not comparable
Granite 4.0 Micro: —, Qwen3-30B-A3B: 29.8 (#82)
| Benchmark | Granite 4.0 Micro | Qwen3-30B-A3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 41.4% |
Reasoning Qwen3-30B-A3B leads
Granite 4.0 Micro: 19.2 (#265), Qwen3-30B-A3B: 22.2 (#204)
| Benchmark | Granite 4.0 Micro | Qwen3-30B-A3B |
|---|---|---|
| Chess Puzzles | 0% | 8% |
| Kagi LLM Benchmark | — | 54.9% |
| CritPt | — | 0.3% |
| LMArena Hard Prompts | — | 1398 |
| DTBench | — | 69.3% |
| LMCA | — | 22.4% |
| Epoch Capabilities Index | — | 139.63 |
Math Qwen3-30B-A3B leads
Granite 4.0 Micro: 12.0 (#307), Qwen3-30B-A3B: 37.4 (#157)
| Benchmark | Granite 4.0 Micro | Qwen3-30B-A3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.8% | 70.3% |
| MathArena Final-Answer Competitions | — | 47.8% |
| Omni-MATH | 20.9% | — |
| LMArena Math | — | 1394 |
Knowledge Qwen3-30B-A3B leads
Granite 4.0 Micro: 9.9 (#304), Qwen3-30B-A3B: 41.8 (#105)
| Benchmark | Granite 4.0 Micro | Qwen3-30B-A3B |
|---|---|---|
| GPQA Diamond | 28.3% | 70.1% |
| MMLU-Pro | 39.5% | — |
| Confabulations | — | 12.3% |
| GPQA (HELM) | 30.7% | — |
| LMArena Expert | — | 1396 |
Multilingual Not comparable
Granite 4.0 Micro: —, Qwen3-30B-A3B: 49.5 (#132)
| Benchmark | Granite 4.0 Micro | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | — | 1372 |
| LMArena Chinese | — | 1433 |
| LMArena French | — | 1418 |
| LMArena German | — | 1380 |
| LMArena Japanese | — | 1337 |
| LMArena Korean | — | 1331 |
| LMArena Russian | — | 1370 |
| LMArena Spanish | — | 1404 |
Instruction Following Qwen3-30B-A3B leads
Granite 4.0 Micro: 69.9 (#169), Qwen3-30B-A3B: 72.0 (#142)
| Benchmark | Granite 4.0 Micro | Qwen3-30B-A3B |
|---|---|---|
| IFEval | 84.9% | — |
| LMArena Instruction Following | — | 1363 |
Long Context Not comparable
Granite 4.0 Micro: —, Qwen3-30B-A3B: 31.0 (#283)
| Benchmark | Granite 4.0 Micro | Qwen3-30B-A3B |
|---|---|---|
| Fiction.LiveBench | — | 40.6% |
| LMArena Longer Query | — | 1379 |
Writing & Preference Qwen3-30B-A3B leads
Granite 4.0 Micro: 46.7 (#216), Qwen3-30B-A3B: 55.6 (#143)
| Benchmark | Granite 4.0 Micro | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | — | 1384 |
| LMArena Creative Writing | — | 1317 |
| Short-Story Creative Writing | — | 75.3% |
| WildBench | 67% | — |
| LMArena Multi-Turn | — | 1378 |
Frequently asked questions
Is Granite 4.0 Micro better than Qwen3-30B-A3B?
Qwen3-30B-A3B is the stronger model overall, scoring 38.9 to 29.0 on the Noometry Index. Granite 4.0 Micro costs 5.3× less per token, which makes it the better buy when Qwen3-30B-A3B's lead doesn't matter for your workload.
Which is cheaper, Granite 4.0 Micro or Qwen3-30B-A3B?
Granite 4.0 Micro is cheaper. It lists at $0.017 per million input tokens and $0.11 per million output tokens; Qwen3-30B-A3B lists at $0.12 and $0.50.
Which has the bigger context window?
Granite 4.0 Micro does, with 131K tokens against 41K.
How many benchmarks do Granite 4.0 Micro and Qwen3-30B-A3B share?
3 benchmarks have published results for both models. Granite 4.0 Micro has 8 scored results on Noometry and Qwen3-30B-A3B has 32.