Model comparison
Gemma 3 12B vs Qwen3.5 27B
Qwen3.5 27B is the stronger model overall, scoring 41.4 to 31.8 on the Noometry Index. Gemma 3 12B costs 11× less per token, which makes it the better buy when Qwen3.5 27B's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 15 benchmarks with published results for both. Gemma 3 12B scores higher in 0 categories and Qwen3.5 27B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.5 27B leads 36.8 to 21.0.
- The biggest single-benchmark swing is DTBench: 48.8% for Gemma 3 12B and 82.4% for Qwen3.5 27B.
- Gemma 3 12B is cheaper at $0.05 / $0.15 per million input/output tokens, against $0.30 / $2.40 for Qwen3.5 27B.
- Qwen3.5 27B accepts more context: 262K tokens versus 131K.
Side by side
| Gemma 3 12B | Qwen3.5 27B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 31.8 | 41.4 |
| Released | 2025-03-12 | 2026-02-23 |
| Weights | Open | Open |
| Context window | 131K | 262K |
| Max output | 8K | 66K |
| Input $ / M tokens | $0.05 | $0.30 |
| Output $ / M tokens | $0.15 | $2.40 |
| Results tracked | 24 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5 27B leads
Gemma 3 12B: 31.2 (#276), Qwen3.5 27B: 38.6 (#159)
| Benchmark | Gemma 3 12B | Qwen3.5 27B |
|---|---|---|
| LMArena Coding | 1281 | 1427 |
| LMArena WebDev | — | 1357 |
| SciCode | 17.4% | — |
| WeirdML | — | 39.5% |
| ALE-Bench | — | 349.45 |
Agentic & Tool Use Not comparable
Gemma 3 12B: 25.3 (#101), Qwen3.5 27B: —
| Benchmark | Gemma 3 12B | Qwen3.5 27B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 30.4% | — |
| Vending-Bench 2 | — | 201.98 |
Reasoning Qwen3.5 27B leads
Gemma 3 12B: 15.2 (#307), Qwen3.5 27B: 26.6 (#116)
| Benchmark | Gemma 3 12B | Qwen3.5 27B |
|---|---|---|
| LMArena Hard Prompts | 1309 | 1414 |
| DTBench | 48.8% | 82.4% |
| LMCA | 4.5% | 34% |
| NYT Connections (extended) | — | 47.9% |
| CritPt | 0% | — |
| Chess Puzzles | 0% | — |
| Thematic Generalization | — | 45.5% |
| Epoch Capabilities Index | 123.49 | — |
Math Qwen3.5 27B leads
Gemma 3 12B: 21.0 (#273), Qwen3.5 27B: 36.8 (#103)
| Benchmark | Gemma 3 12B | Qwen3.5 27B |
|---|---|---|
| LMArena Math | 1307 | 1428 |
| MathArena Final-Answer Competitions | — | 56.7% |
| OTIS Mock AIME 2024-2025 | 16.7% | — |
Knowledge Qwen3.5 27B leads
Gemma 3 12B: 26.5 (#250), Qwen3.5 27B: 37.4 (#149)
| Benchmark | Gemma 3 12B | Qwen3.5 27B |
|---|---|---|
| Vectara Hallucination Rate | 4.4% | 12.1% |
| LMArena Expert | 1248 | 1428 |
| GPQA Diamond | 39.5% | — |
Multimodal Not comparable
Gemma 3 12B: —, Qwen3.5 27B: 38.7 (#56)
| Benchmark | Gemma 3 12B | Qwen3.5 27B |
|---|---|---|
| LMArena Vision | — | 1240 |
| MindCube | 46.7% | — |
Multilingual Qwen3.5 27B leads
Gemma 3 12B: 45.9 (#160), Qwen3.5 27B: 51.0 (#112)
| Benchmark | Gemma 3 12B | Qwen3.5 27B |
|---|---|---|
| LMArena Non-English | 1318 | 1390 |
| LMArena German | 1370 | 1393 |
| LMArena Russian | 1335 | 1389 |
| LMArena Chinese | — | 1478 |
| LMArena French | — | 1409 |
| LMArena Japanese | — | 1345 |
| LMArena Korean | — | 1359 |
| LMArena Spanish | — | 1408 |
Instruction Following Qwen3.5 27B leads
Gemma 3 12B: 68.8 (#179), Qwen3.5 27B: 73.7 (#114)
| Benchmark | Gemma 3 12B | Qwen3.5 27B |
|---|---|---|
| LMArena Instruction Following | 1299 | 1393 |
Long Context Qwen3.5 27B leads
Gemma 3 12B: 40.0 (#157), Qwen3.5 27B: 43.1 (#103)
| Benchmark | Gemma 3 12B | Qwen3.5 27B |
|---|---|---|
| LMArena Longer Query | 1317 | 1413 |
Writing & Preference Qwen3.5 27B leads
Gemma 3 12B: 47.7 (#205), Qwen3.5 27B: 59.5 (#106)
| Benchmark | Gemma 3 12B | Qwen3.5 27B |
|---|---|---|
| LMArena Text | 1334 | 1409 |
| LMArena Creative Writing | 1331 | 1362 |
| LMArena Multi-Turn | 1334 | 1410 |
| EQ-Bench Creative Writing | 1126 | — |
Frequently asked questions
Is Gemma 3 12B better than Qwen3.5 27B?
Qwen3.5 27B is the stronger model overall, scoring 41.4 to 31.8 on the Noometry Index. Gemma 3 12B costs 11× less per token, which makes it the better buy when Qwen3.5 27B's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 12B or Qwen3.5 27B?
Gemma 3 12B is cheaper. It lists at $0.05 per million input tokens and $0.15 per million output tokens; Qwen3.5 27B lists at $0.30 and $2.40.
Is Gemma 3 12B or Qwen3.5 27B better for coding?
Qwen3.5 27B scores higher on coding benchmarks: 38.6 versus 31.2 in the Noometry coding category.
Which has the bigger context window?
Qwen3.5 27B does, with 262K tokens against 131K.
How many benchmarks do Gemma 3 12B and Qwen3.5 27B share?
15 benchmarks have published results for both models. Gemma 3 12B has 24 scored results on Noometry and Qwen3.5 27B has 28.