Model comparison
Gemma 4 31B IT vs Qwen3.8 27B
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 43.5 on the Noometry Index. Gemma 4 31B IT costs 4.0× 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 4 31B IT scores higher in 3 categories and Qwen3.8 27B in 6 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 27B leads 41.0 to 27.2.
- The biggest single-benchmark swing is NYT Connections (extended): 70.6% for Gemma 4 31B IT and 54.5% for Qwen3.8 27B.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $0.04 / $2.30 for Qwen3.8 27B.
Side by side
| Gemma 4 31B IT | Qwen3.8 27B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 43.5 | 46.0 |
| Released | 2026-04-02 | 2026-08-14 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 33K | 33K |
| Input $ / M tokens | $0.09 | $0.04 |
| Output $ / M tokens | $0.34 | $2.30 |
| Results tracked | 35 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
Gemma 4 31B IT: 42.3 (#108), Qwen3.8 27B: 50.5 (#44)
| Benchmark | Gemma 4 31B IT | Qwen3.8 27B |
|---|---|---|
| LMArena WebDev | 1366 | 1593 |
| SciCode | 43.4% | 46.6% |
| LMArena Coding | 1459 | 1482 |
| WeirdML | 52.3% | — |
| ALE-Bench | 925.5 | — |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, Qwen3.8 27B: 32.9 (#57)
| Benchmark | Gemma 4 31B IT | Qwen3.8 27B |
|---|---|---|
| APEX-Agents | — | 47.5% |
Reasoning Qwen3.8 27B leads
Gemma 4 31B IT: 27.2 (#122), Qwen3.8 27B: 41.0 (#54)
| Benchmark | Gemma 4 31B IT | Qwen3.8 27B |
|---|---|---|
| NYT Connections (extended) | 70.6% | 54.5% |
| CritPt | 1.4% | 5.4% |
| LMArena Hard Prompts | 1448 | 1460 |
| DTBench | 82.7% | 88% |
| LMCA | 39.3% | 41.4% |
| Surface Evolver Bench | 30.6% | 45% |
| Epoch Capabilities Index | 142.74 | 149.38 |
| ARC-AGI-2 | — | 42.4% |
| Kagi LLM Benchmark | 63.5% | — |
| ARC-AGI-1 | — | 87.5% |
| Chess Puzzles | 5% | — |
| Thematic Generalization | 53% | — |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 43.2 (#81), Qwen3.8 27B: 37.1 (#161)
| Benchmark | Gemma 4 31B IT | Qwen3.8 27B |
|---|---|---|
| LMArena Math | 1465 | 1456 |
| OTIS Mock AIME 2024-2025 | 73.3% | — |
| ProofBench | — | 16% |
Knowledge Qwen3.8 27B leads
Gemma 4 31B IT: 37.9 (#151), Qwen3.8 27B: 41.6 (#109)
| Benchmark | Gemma 4 31B IT | Qwen3.8 27B |
|---|---|---|
| LMArena Expert | 1465 | 1482 |
| GPQA Diamond | 75.8% | — |
| SimpleQA Verified | 10.4% | — |
| Vectara Hallucination Rate | 7.4% | — |
Multimodal Too close to call
Gemma 4 31B IT: 41.6 (#34), Qwen3.8 27B: 41.3 (#37)
| Benchmark | Gemma 4 31B IT | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | 1277 | 1271 |
| LMArena Document | 1425 | — |
Multilingual Too close to call
Gemma 4 31B IT: 53.8 (#57), Qwen3.8 27B: 53.7 (#60)
| Benchmark | Gemma 4 31B IT | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | 1431 | 1430 |
| LMArena Chinese | 1476 | 1504 |
| LMArena French | 1435 | 1465 |
| LMArena Russian | 1460 | 1415 |
| LMArena Spanish | 1444 | 1448 |
| LMArena German | — | 1438 |
| LMArena Japanese | — | 1384 |
| LMArena Korean | — | 1393 |
Instruction Following Too close to call
Gemma 4 31B IT: 75.5 (#61), Qwen3.8 27B: 75.8 (#53)
| Benchmark | Gemma 4 31B IT | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | 1433 | 1439 |
Long Context Too close to call
Gemma 4 31B IT: 44.2 (#71), Qwen3.8 27B: 44.3 (#70)
| Benchmark | Gemma 4 31B IT | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | 1446 | 1450 |
Writing & Preference Qwen3.8 27B leads
Gemma 4 31B IT: 60.5 (#96), Qwen3.8 27B: 65.8 (#43)
| Benchmark | Gemma 4 31B IT | Qwen3.8 27B |
|---|---|---|
| LMArena Text | 1443 | 1441 |
| LMArena Creative Writing | 1415 | 1384 |
| EQ-Bench Creative Writing | 1368 | 1671 |
| LMArena Multi-Turn | 1452 | 1441 |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than Qwen3.8 27B?
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 43.5 on the Noometry Index. Gemma 4 31B IT costs 4.0× 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 4 31B IT or Qwen3.8 27B?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; Qwen3.8 27B lists at $0.04 and $2.30.
Is Gemma 4 31B IT or Qwen3.8 27B better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 42.3 in the Noometry coding category.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Gemma 4 31B IT and Qwen3.8 27B share?
24 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and Qwen3.8 27B has 31.