Model comparison
Gemma 3 27B vs Qwen3.8 27B
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 30.8 on the Noometry Index. Gemma 3 27B costs 6.0× less per token, which makes it the better buy when Qwen3.8 27B's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. Gemma 3 27B scores higher in 0 categories and Qwen3.8 27B in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in coding, where Qwen3.8 27B leads 50.5 to 22.5.
- The biggest single-benchmark swing is DTBench: 52.5% for Gemma 3 27B and 88% for Qwen3.8 27B.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $0.04 / $2.30 for Qwen3.8 27B.
- Qwen3.8 27B accepts more context: 262K tokens versus 131K.
Side by side
| Gemma 3 27B | Qwen3.8 27B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 30.8 | 46.0 |
| Released | 2025-03-11 | 2026-08-14 |
| Weights | Open | Open |
| Context window | 131K | 262K |
| Max output | 8K | 33K |
| Input $ / M tokens | $0.08 | $0.04 |
| Output $ / M tokens | $0.16 | $2.30 |
| Results tracked | 43 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
Gemma 3 27B: 22.5 (#334), Qwen3.8 27B: 50.5 (#44)
| Benchmark | Gemma 3 27B | Qwen3.8 27B |
|---|---|---|
| SciCode | 21.2% | 46.6% |
| LMArena Coding | 1322 | 1482 |
| Aider Polyglot | 4.9% | — |
| LMArena WebDev | — | 1593 |
| LiveBench Coding | 39.9% | — |
Agentic & Tool Use Qwen3.8 27B leads
Gemma 3 27B: 25.1 (#110), Qwen3.8 27B: 32.9 (#57)
| Benchmark | Gemma 3 27B | Qwen3.8 27B |
|---|---|---|
| APEX-Agents | — | 47.5% |
| Berkeley Function Calling Leaderboard | 29.5% | — |
Reasoning Qwen3.8 27B leads
Gemma 3 27B: 16.7 (#301), Qwen3.8 27B: 41.0 (#54)
| Benchmark | Gemma 3 27B | Qwen3.8 27B |
|---|---|---|
| CritPt | 0% | 5.4% |
| LMArena Hard Prompts | 1340 | 1460 |
| DTBench | 52.5% | 88% |
| LMCA | 12.3% | 41.4% |
| Epoch Capabilities Index | 130.04 | 149.38 |
| ARC-AGI-2 | — | 42.4% |
| Kagi LLM Benchmark | 40.4% | — |
| NYT Connections (extended) | — | 54.5% |
| ARC-AGI-1 | — | 87.5% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | 43.8% | — |
| LiveBench Data Analysis | 51.5% | — |
| Surface Evolver Bench | — | 45% |
| LiveBench | 50% | — |
Math Qwen3.8 27B leads
Gemma 3 27B: 25.9 (#265), Qwen3.8 27B: 37.1 (#161)
| Benchmark | Gemma 3 27B | Qwen3.8 27B |
|---|---|---|
| LMArena Math | 1312 | 1456 |
| OTIS Mock AIME 2024-2025 | 22.5% | — |
| ProofBench | — | 16% |
| LiveBench Math | 55.4% | — |
| MATH Level 5 | 74% | — |
Knowledge Qwen3.8 27B leads
Gemma 3 27B: 25.5 (#261), Qwen3.8 27B: 41.6 (#109)
| Benchmark | Gemma 3 27B | Qwen3.8 27B |
|---|---|---|
| LMArena Expert | 1304 | 1482 |
| GPQA Diamond | 47.7% | — |
| Confabulations | 40.3% | — |
| Vectara Hallucination Rate | 7.4% | — |
Multimodal Qwen3.8 27B leads
Gemma 3 27B: 32.6 (#100), Qwen3.8 27B: 41.3 (#37)
| Benchmark | Gemma 3 27B | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | 1164 | 1271 |
| GeoBench | 52% | — |
Multilingual Qwen3.8 27B leads
Gemma 3 27B: 46.9 (#155), Qwen3.8 27B: 53.7 (#60)
| Benchmark | Gemma 3 27B | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | 1334 | 1430 |
| LMArena Chinese | 1346 | 1504 |
| LMArena French | 1368 | 1465 |
| LMArena German | 1362 | 1438 |
| LMArena Japanese | 1287 | 1384 |
| LMArena Korean | 1308 | 1393 |
| LMArena Russian | 1349 | 1415 |
| LMArena Spanish | 1349 | 1448 |
Instruction Following Qwen3.8 27B leads
Gemma 3 27B: 70.6 (#160), Qwen3.8 27B: 75.8 (#53)
| Benchmark | Gemma 3 27B | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | 1321 | 1439 |
| LiveBench Instruction Following | 74.9% | — |
Long Context Qwen3.8 27B leads
Gemma 3 27B: 27.6 (#293), Qwen3.8 27B: 44.3 (#70)
| Benchmark | Gemma 3 27B | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | 1333 | 1450 |
| Fiction.LiveBench | 33.3% | — |
Writing & Preference Qwen3.8 27B leads
Gemma 3 27B: 52.5 (#168), Qwen3.8 27B: 65.8 (#43)
| Benchmark | Gemma 3 27B | Qwen3.8 27B |
|---|---|---|
| LMArena Text | 1358 | 1441 |
| LMArena Creative Writing | 1346 | 1384 |
| EQ-Bench Creative Writing | 1266 | 1671 |
| LMArena Multi-Turn | 1345 | 1441 |
| Short-Story Creative Writing | 79.9% | — |
| LiveBench Language | 34.6% | — |
Frequently asked questions
Is Gemma 3 27B better than Qwen3.8 27B?
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 30.8 on the Noometry Index. Gemma 3 27B costs 6.0× less per token, which makes it the better buy when Qwen3.8 27B's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 27B or Qwen3.8 27B?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; Qwen3.8 27B lists at $0.04 and $2.30.
Is Gemma 3 27B or Qwen3.8 27B better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 22.5 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 27B does, with 262K tokens against 131K.
How many benchmarks do Gemma 3 27B and Qwen3.8 27B share?
24 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and Qwen3.8 27B has 31.