Model comparison
DeepSeek-V3.1 vs Gemma 4 31B IT
DeepSeek-V3.1 and Gemma 4 31B IT score almost the same on the Noometry Index (42.8 vs 43.5), so choose on price, context window or the category you care about most.
Last verified . 21 shared benchmarks.
DeepSeek
GoogleSummary
- They share 21 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 2 categories and Gemma 4 31B IT in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in long context, where Gemma 4 31B IT leads 44.2 to 36.3.
- The biggest single-benchmark swing is LMCA: 24.3% for DeepSeek-V3.1 and 39.3% for Gemma 4 31B IT.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $0.25 / $0.95 for DeepSeek-V3.1.
- Gemma 4 31B IT accepts more context: 262K tokens versus 164K.
Side by side
| DeepSeek-V3.1 | Gemma 4 31B IT | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 42.8 | 43.5 |
| Released | 2025-08-21 | 2026-04-02 |
| Weights | Open | Open |
| Context window | 164K | 262K |
| Max output | 8K | 33K |
| Input $ / M tokens | $0.25 | $0.09 |
| Output $ / M tokens | $0.95 | $0.34 |
| Results tracked | 27 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 31B IT leads
DeepSeek-V3.1: 40.3 (#144), Gemma 4 31B IT: 42.3 (#108)
| Benchmark | DeepSeek-V3.1 | Gemma 4 31B IT |
|---|---|---|
| WeirdML | 38.4% | 52.3% |
| LMArena Coding | 1417 | 1459 |
| LMArena WebDev | — | 1366 |
| SciCode | — | 43.4% |
| ALE-Bench | — | 925.5 |
Reasoning Too close to call
DeepSeek-V3.1: 27.9 (#110), Gemma 4 31B IT: 27.2 (#122)
| Benchmark | DeepSeek-V3.1 | Gemma 4 31B IT |
|---|---|---|
| Kagi LLM Benchmark | 53.2% | 63.5% |
| LMArena Hard Prompts | 1417 | 1448 |
| DTBench | 82.7% | 82.7% |
| LMCA | 24.3% | 39.3% |
| Epoch Capabilities Index | 139.92 | 142.74 |
| SimpleBench | 40% | — |
| NYT Connections (extended) | — | 70.6% |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 5% |
| Thematic Generalization | — | 53% |
| Surface Evolver Bench | — | 30.6% |
| ForecastBench | 58 | — |
Math Gemma 4 31B IT leads
DeepSeek-V3.1: 38.9 (#122), Gemma 4 31B IT: 43.2 (#81)
| Benchmark | DeepSeek-V3.1 | Gemma 4 31B IT |
|---|---|---|
| LMArena Math | 1420 | 1465 |
| OTIS Mock AIME 2024-2025 | — | 73.3% |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), Gemma 4 31B IT: 37.9 (#151)
| Benchmark | DeepSeek-V3.1 | Gemma 4 31B IT |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | 7.4% |
| LMArena Expert | 1405 | 1465 |
| GPQA Diamond | — | 75.8% |
| SimpleQA Verified | — | 10.4% |
Multimodal Not comparable
DeepSeek-V3.1: —, Gemma 4 31B IT: 41.6 (#34)
| Benchmark | DeepSeek-V3.1 | Gemma 4 31B IT |
|---|---|---|
| LMArena Vision | — | 1277 |
| LMArena Document | — | 1425 |
Multilingual Gemma 4 31B IT leads
DeepSeek-V3.1: 51.6 (#106), Gemma 4 31B IT: 53.8 (#57)
| Benchmark | DeepSeek-V3.1 | Gemma 4 31B IT |
|---|---|---|
| LMArena Non-English | 1400 | 1431 |
| LMArena Chinese | 1469 | 1476 |
| LMArena French | 1447 | 1435 |
| LMArena Russian | 1405 | 1460 |
| LMArena Spanish | 1431 | 1444 |
| LMArena German | 1411 | — |
| LMArena Japanese | 1378 | — |
| LMArena Korean | 1337 | — |
Instruction Following Gemma 4 31B IT leads
DeepSeek-V3.1: 73.9 (#110), Gemma 4 31B IT: 75.5 (#61)
| Benchmark | DeepSeek-V3.1 | Gemma 4 31B IT |
|---|---|---|
| LMArena Instruction Following | 1400 | 1433 |
Long Context Gemma 4 31B IT leads
DeepSeek-V3.1: 36.3 (#232), Gemma 4 31B IT: 44.2 (#71)
| Benchmark | DeepSeek-V3.1 | Gemma 4 31B IT |
|---|---|---|
| LMArena Longer Query | 1422 | 1446 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference Too close to call
DeepSeek-V3.1: 60.3 (#98), Gemma 4 31B IT: 60.5 (#96)
| Benchmark | DeepSeek-V3.1 | Gemma 4 31B IT |
|---|---|---|
| LMArena Text | 1420 | 1443 |
| LMArena Creative Writing | 1401 | 1415 |
| EQ-Bench Creative Writing | 1436 | 1368 |
| LMArena Multi-Turn | 1408 | 1452 |
| EQ-Bench 4 | — | 1120 |
Frequently asked questions
Is DeepSeek-V3.1 better than Gemma 4 31B IT?
DeepSeek-V3.1 and Gemma 4 31B IT score almost the same on the Noometry Index (42.8 vs 43.5), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-V3.1 or Gemma 4 31B IT?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; DeepSeek-V3.1 lists at $0.25 and $0.95.
Is DeepSeek-V3.1 or Gemma 4 31B IT better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 42.3 versus 40.3 in the Noometry coding category.
Which has the bigger context window?
Gemma 4 31B IT does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-V3.1 and Gemma 4 31B IT share?
21 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Gemma 4 31B IT has 35.