Model comparison
Gemma 3 27B vs Qwen3-30B-A3B
Qwen3-30B-A3B is the stronger model overall, scoring 38.9 to 30.8 on the Noometry Index. Gemma 3 27B costs 2.1× less per token, which makes it the better buy when Qwen3-30B-A3B's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 30 benchmarks with published results for both. Gemma 3 27B scores higher in 0 categories and Qwen3-30B-A3B in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3-30B-A3B leads 41.8 to 25.5.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 22.5% for Gemma 3 27B and 70.3% for Qwen3-30B-A3B.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $0.12 / $0.50 for Qwen3-30B-A3B.
- Gemma 3 27B accepts more context: 131K tokens versus 41K.
Side by side
| Gemma 3 27B | Qwen3-30B-A3B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 30.8 | 38.9 |
| Released | 2025-03-11 | 2025-04-28 |
| Weights | Open | Open |
| Context window | 131K | 41K |
| Max output | 8K | 16K |
| Input $ / M tokens | $0.08 | $0.12 |
| Output $ / M tokens | $0.16 | $0.50 |
| Results tracked | 43 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3-30B-A3B leads
Gemma 3 27B: 22.5 (#334), Qwen3-30B-A3B: 37.5 (#194)
| Benchmark | Gemma 3 27B | Qwen3-30B-A3B |
|---|---|---|
| SciCode | 21.2% | 33.3% |
| LMArena Coding | 1322 | 1416 |
| Aider Polyglot | 4.9% | — |
| WeirdML | — | 29.8% |
| LiveBench Coding | 39.9% | — |
Agentic & Tool Use Qwen3-30B-A3B leads
Gemma 3 27B: 25.1 (#110), Qwen3-30B-A3B: 29.8 (#82)
| Benchmark | Gemma 3 27B | Qwen3-30B-A3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 29.5% | 41.4% |
Reasoning Qwen3-30B-A3B leads
Gemma 3 27B: 16.7 (#301), Qwen3-30B-A3B: 22.2 (#204)
| Benchmark | Gemma 3 27B | Qwen3-30B-A3B |
|---|---|---|
| Kagi LLM Benchmark | 40.4% | 54.9% |
| CritPt | 0% | 0.3% |
| Chess Puzzles | 0% | 8% |
| LMArena Hard Prompts | 1340 | 1398 |
| DTBench | 52.5% | 69.3% |
| LMCA | 12.3% | 22.4% |
| Epoch Capabilities Index | 130.04 | 139.63 |
| LiveBench Reasoning | 43.8% | — |
| LiveBench Data Analysis | 51.5% | — |
| LiveBench | 50% | — |
Math Qwen3-30B-A3B leads
Gemma 3 27B: 25.9 (#265), Qwen3-30B-A3B: 37.4 (#157)
| Benchmark | Gemma 3 27B | Qwen3-30B-A3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 22.5% | 70.3% |
| LMArena Math | 1312 | 1394 |
| MathArena Final-Answer Competitions | — | 47.8% |
| LiveBench Math | 55.4% | — |
| MATH Level 5 | 74% | — |
Knowledge Qwen3-30B-A3B leads
Gemma 3 27B: 25.5 (#261), Qwen3-30B-A3B: 41.8 (#105)
| Benchmark | Gemma 3 27B | Qwen3-30B-A3B |
|---|---|---|
| GPQA Diamond | 47.7% | 70.1% |
| Confabulations | 40.3% | 12.3% |
| LMArena Expert | 1304 | 1396 |
| Vectara Hallucination Rate | 7.4% | — |
Multimodal Not comparable
Gemma 3 27B: 32.6 (#100), Qwen3-30B-A3B: —
| Benchmark | Gemma 3 27B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Vision | 1164 | — |
| GeoBench | 52% | — |
Multilingual Qwen3-30B-A3B leads
Gemma 3 27B: 46.9 (#155), Qwen3-30B-A3B: 49.5 (#132)
| Benchmark | Gemma 3 27B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | 1334 | 1372 |
| LMArena Chinese | 1346 | 1433 |
| LMArena French | 1368 | 1418 |
| LMArena German | 1362 | 1380 |
| LMArena Japanese | 1287 | 1337 |
| LMArena Korean | 1308 | 1331 |
| LMArena Russian | 1349 | 1370 |
| LMArena Spanish | 1349 | 1404 |
Instruction Following Qwen3-30B-A3B leads
Gemma 3 27B: 70.6 (#160), Qwen3-30B-A3B: 72.0 (#142)
| Benchmark | Gemma 3 27B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Instruction Following | 1321 | 1363 |
| LiveBench Instruction Following | 74.9% | — |
Long Context Qwen3-30B-A3B leads
Gemma 3 27B: 27.6 (#293), Qwen3-30B-A3B: 31.0 (#283)
| Benchmark | Gemma 3 27B | Qwen3-30B-A3B |
|---|---|---|
| Fiction.LiveBench | 33.3% | 40.6% |
| LMArena Longer Query | 1333 | 1379 |
Writing & Preference Qwen3-30B-A3B leads
Gemma 3 27B: 52.5 (#168), Qwen3-30B-A3B: 55.6 (#143)
| Benchmark | Gemma 3 27B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | 1358 | 1384 |
| LMArena Creative Writing | 1346 | 1317 |
| Short-Story Creative Writing | 79.9% | 75.3% |
| LMArena Multi-Turn | 1345 | 1378 |
| EQ-Bench Creative Writing | 1266 | — |
| LiveBench Language | 34.6% | — |
Frequently asked questions
Is Gemma 3 27B better than Qwen3-30B-A3B?
Qwen3-30B-A3B is the stronger model overall, scoring 38.9 to 30.8 on the Noometry Index. Gemma 3 27B costs 2.1× less per token, which makes it the better buy when Qwen3-30B-A3B's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 27B or Qwen3-30B-A3B?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; Qwen3-30B-A3B lists at $0.12 and $0.50.
Is Gemma 3 27B or Qwen3-30B-A3B better for coding?
Qwen3-30B-A3B scores higher on coding benchmarks: 37.5 versus 22.5 in the Noometry coding category.
Which has the bigger context window?
Gemma 3 27B does, with 131K tokens against 41K.
How many benchmarks do Gemma 3 27B and Qwen3-30B-A3B share?
30 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and Qwen3-30B-A3B has 32.