Model comparison
Gemma 2 27B vs Qwen3.5 397B-A17B
Qwen3.5 397B-A17B is the stronger model overall, scoring 46.0 to 29.4 on the Noometry Index. Gemma 2 27B costs 2.1× less per token, which makes it the better buy when Qwen3.5 397B-A17B's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 22 benchmarks with published results for both. Gemma 2 27B scores higher in 0 categories and Qwen3.5 397B-A17B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.5 397B-A17B leads 46.1 to 10.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.4% for Gemma 2 27B and 88.9% for Qwen3.5 397B-A17B.
- Gemma 2 27B is cheaper at $0.65 / $0.65 per million input/output tokens, against $0.60 / $3.60 for Qwen3.5 397B-A17B.
- Qwen3.5 397B-A17B accepts more context: 262K tokens versus 8K.
Side by side
| Gemma 2 27B | Qwen3.5 397B-A17B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 29.4 | 46.0 |
| Released | 2024-06-24 | 2026-02-01 |
| Weights | Open | Open |
| Context window | 8K | 262K |
| Max output | 2K | 66K |
| Input $ / M tokens | $0.65 | $0.60 |
| Output $ / M tokens | $0.65 | $3.60 |
| Results tracked | 34 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5 397B-A17B leads
Gemma 2 27B: 34.1 (#246), Qwen3.5 397B-A17B: 42.0 (#114)
| Benchmark | Gemma 2 27B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Coding | 1211 | 1465 |
| LMArena WebDev | — | 1400 |
| BigCodeBench Instruct | 42.8% | — |
| LiveBench Coding | 36% | — |
| BigCodeBench Complete | 52.5% | — |
Agentic & Tool Use Not comparable
Gemma 2 27B: —, Qwen3.5 397B-A17B: 33.3 (#53)
| Benchmark | Gemma 2 27B | Qwen3.5 397B-A17B |
|---|---|---|
| APEX-Agents | — | 24.9% |
| τ²-bench Airline | — | 81.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 84.4% |
| τ²-bench Telecom | — | 97.8% |
Reasoning Qwen3.5 397B-A17B leads
Gemma 2 27B: 15.3 (#315), Qwen3.5 397B-A17B: 34.5 (#70)
| Benchmark | Gemma 2 27B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Hard Prompts | 1198 | 1448 |
| DTBench | 48% | 87.5% |
| LMCA | 7.1% | 37.9% |
| Epoch Capabilities Index | 122.08 | 146.65 |
| Kagi LLM Benchmark | — | 73.7% |
| NYT Connections (extended) | — | 58.9% |
| Chess Puzzles | — | 13% |
| Thematic Generalization | — | 65.1% |
| LiveBench Reasoning | 28.1% | — |
| Mystery Game Puzzles | — | 18% |
| LiveBench Data Analysis | 47.9% | — |
| LiveBench | 38.2% | — |
Math Qwen3.5 397B-A17B leads
Gemma 2 27B: 10.7 (#311), Qwen3.5 397B-A17B: 46.1 (#73)
| Benchmark | Gemma 2 27B | Qwen3.5 397B-A17B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.4% | 88.9% |
| LMArena Math | 1212 | 1454 |
| FrontierMath (Tiers 1-3) | — | 31.2% |
| LiveBench Math | 26.5% | — |
| MATH Level 5 | 27.9% | — |
Knowledge Qwen3.5 397B-A17B leads
Gemma 2 27B: 19.0 (#280), Qwen3.5 397B-A17B: 53.3 (#58)
| Benchmark | Gemma 2 27B | Qwen3.5 397B-A17B |
|---|---|---|
| GPQA Diamond | 36.5% | 86.4% |
| LMArena Expert | 1172 | 1462 |
| Confabulations | 27.1% | — |
| MMLU | 75.7% | — |
Multimodal Not comparable
Gemma 2 27B: —, Qwen3.5 397B-A17B: 40.7 (#44)
| Benchmark | Gemma 2 27B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Vision | — | 1263 |
Multilingual Qwen3.5 397B-A17B leads
Gemma 2 27B: 38.6 (#226), Qwen3.5 397B-A17B: 53.7 (#59)
| Benchmark | Gemma 2 27B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Non-English | 1217 | 1430 |
| LMArena Chinese | 1221 | 1500 |
| LMArena French | 1247 | 1461 |
| LMArena German | 1209 | 1447 |
| LMArena Japanese | 1175 | 1426 |
| LMArena Korean | 1174 | 1384 |
| LMArena Russian | 1234 | 1429 |
| LMArena Spanish | 1228 | 1441 |
Instruction Following Qwen3.5 397B-A17B leads
Gemma 2 27B: 60.5 (#249), Qwen3.5 397B-A17B: 75.0 (#77)
| Benchmark | Gemma 2 27B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Instruction Following | 1206 | 1424 |
| LiveBench Instruction Following | 58.1% | — |
Long Context Qwen3.5 397B-A17B leads
Gemma 2 27B: 37.3 (#218), Qwen3.5 397B-A17B: 44.1 (#74)
| Benchmark | Gemma 2 27B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Longer Query | 1231 | 1442 |
Writing & Preference Qwen3.5 397B-A17B leads
Gemma 2 27B: 44.2 (#225), Qwen3.5 397B-A17B: 62.3 (#79)
| Benchmark | Gemma 2 27B | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Text | 1231 | 1438 |
| LMArena Creative Writing | 1241 | 1401 |
| LMArena Multi-Turn | 1224 | 1446 |
| EQ-Bench Creative Writing | — | 1478 |
| LiveBench Language | 32.6% | — |
Frequently asked questions
Is Gemma 2 27B better than Qwen3.5 397B-A17B?
Qwen3.5 397B-A17B is the stronger model overall, scoring 46.0 to 29.4 on the Noometry Index. Gemma 2 27B costs 2.1× less per token, which makes it the better buy when Qwen3.5 397B-A17B's lead doesn't matter for your workload.
Which is cheaper, Gemma 2 27B or Qwen3.5 397B-A17B?
Gemma 2 27B is cheaper. It lists at $0.65 per million input tokens and $0.65 per million output tokens; Qwen3.5 397B-A17B lists at $0.60 and $3.60.
Is Gemma 2 27B or Qwen3.5 397B-A17B better for coding?
Qwen3.5 397B-A17B scores higher on coding benchmarks: 42.0 versus 34.1 in the Noometry coding category.
Which has the bigger context window?
Qwen3.5 397B-A17B does, with 262K tokens against 8K.
How many benchmarks do Gemma 2 27B and Qwen3.5 397B-A17B share?
22 benchmarks have published results for both models. Gemma 2 27B has 34 scored results on Noometry and Qwen3.5 397B-A17B has 36.