# DeepSeek V4.1 Flash vs Gemma 4 31B IT

> DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 43.5 on the Noometry Index. Gemma 4 31B IT costs 1.7× less per token, which makes it the better buy when DeepSeek V4.1 Flash's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/deepseek-v4-1-flash-vs-gemma-4-31b-it
- Last updated: 2026-10-12
- Shared benchmarks: 27

## Summary

- They share 27 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 8 categories and Gemma 4 31B IT in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4.1 Flash leads 66.7 to 43.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.3% for DeepSeek V4.1 Flash and 73.3% for Gemma 4 31B IT.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $0.15 / $0.60 for DeepSeek V4.1 Flash.
- DeepSeek V4.1 Flash accepts more context: 1M tokens versus 262K.

## Snapshot

| | DeepSeek V4.1 Flash | Gemma 4 31B IT |
|---|---|---|
| Provider | DeepSeek | Google |
| Noometry Index | 52.8 | 43.5 |
| Rank | 38 | 90 |
| Context | 1M | 262K |
| Input $/M | $0.15 | $0.09 |
| Output $/M | $0.60 | $0.34 |
| Weights | Open | Open |

## Coding

- DeepSeek V4.1 Flash: 52.9 (#32)
- Gemma 4 31B IT: 42.3 (#108)

| Benchmark | DeepSeek V4.1 Flash | Gemma 4 31B IT |
|---|---|---|
| LMArena WebDev | 1619 | 1366 |
| SciCode | 51.9% | 43.4% |
| LMArena Coding | 1506 | 1459 |
| ALE-Bench | 1,092 | 925.5 |
| WeirdML | — | 52.3% |

## Agentic & Tool Use

- DeepSeek V4.1 Flash: 31.2 (#69)
- Gemma 4 31B IT: —

| Benchmark | DeepSeek V4.1 Flash | Gemma 4 31B IT |
|---|---|---|
| APEX-Agents | 39.5% | — |
| GDP.pdf | 19.8% | — |

## Reasoning

- DeepSeek V4.1 Flash: 50.2 (#36)
- Gemma 4 31B IT: 27.2 (#122)

| Benchmark | DeepSeek V4.1 Flash | Gemma 4 31B IT |
|---|---|---|
| NYT Connections (extended) | 89.6% | 70.6% |
| CritPt | 14.3% | 1.4% |
| LMArena Hard Prompts | 1483 | 1448 |
| DTBench | 89.9% | 82.7% |
| LMCA | 47% | 39.3% |
| Surface Evolver Bench | 46.3% | 30.6% |
| Epoch Capabilities Index | 154.9 | 142.74 |
| Kagi LLM Benchmark | — | 63.5% |
| Chess Puzzles | — | 5% |
| Thematic Generalization | — | 53% |
| Mystery Game Puzzles | 43% | — |

## Math

- DeepSeek V4.1 Flash: 66.7 (#25)
- Gemma 4 31B IT: 43.2 (#81)

| Benchmark | DeepSeek V4.1 Flash | Gemma 4 31B IT |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.3% | 73.3% |
| LMArena Math | 1477 | 1465 |
| FrontierMath (Tiers 1-3) | 67.4% | — |
| FrontierMath Tier 4 | 26.8% | — |
| ProofBench | 54% | — |

## Knowledge

- DeepSeek V4.1 Flash: 57.9 (#38)
- Gemma 4 31B IT: 37.9 (#151)

| Benchmark | DeepSeek V4.1 Flash | Gemma 4 31B IT |
|---|---|---|
| GPQA Diamond | 89.8% | 75.8% |
| LMArena Expert | 1506 | 1465 |
| SimpleQA Verified | — | 10.4% |
| Vectara Hallucination Rate | — | 7.4% |

## Multimodal

- DeepSeek V4.1 Flash: 39.1 (#61)
- Gemma 4 31B IT: 41.6 (#34)

| Benchmark | DeepSeek V4.1 Flash | Gemma 4 31B IT |
|---|---|---|
| LMArena Vision | 1277 | 1277 |
| Furniture Assembly | 34.2% | — |
| LMArena Document | — | 1425 |

## Multilingual

- DeepSeek V4.1 Flash: 55.0 (#35)
- Gemma 4 31B IT: 53.8 (#57)

| Benchmark | DeepSeek V4.1 Flash | Gemma 4 31B IT |
|---|---|---|
| LMArena Non-English | 1448 | 1431 |
| LMArena Chinese | 1497 | 1476 |
| LMArena French | 1452 | 1435 |
| LMArena Russian | 1471 | 1460 |
| LMArena Spanish | 1459 | 1444 |
| LMArena German | 1484 | — |
| LMArena Japanese | 1412 | — |
| LMArena Korean | 1452 | — |

## Instruction Following

- DeepSeek V4.1 Flash: 77.3 (#26)
- Gemma 4 31B IT: 75.5 (#61)

| Benchmark | DeepSeek V4.1 Flash | Gemma 4 31B IT |
|---|---|---|
| LMArena Instruction Following | 1474 | 1433 |

## Long Context

- DeepSeek V4.1 Flash: 45.2 (#47)
- Gemma 4 31B IT: 44.2 (#71)

| Benchmark | DeepSeek V4.1 Flash | Gemma 4 31B IT |
|---|---|---|
| LMArena Longer Query | 1475 | 1446 |

## Writing & Preference

- DeepSeek V4.1 Flash: 65.4 (#48)
- Gemma 4 31B IT: 60.5 (#96)

| Benchmark | DeepSeek V4.1 Flash | Gemma 4 31B IT |
|---|---|---|
| LMArena Text | 1462 | 1443 |
| LMArena Creative Writing | 1435 | 1415 |
| EQ-Bench Creative Writing | 1540 | 1368 |
| LMArena Multi-Turn | 1457 | 1452 |
| EQ-Bench 4 | — | 1120 |

## FAQ

### Is DeepSeek V4.1 Flash better than Gemma 4 31B IT?

DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 43.5 on the Noometry Index. Gemma 4 31B IT costs 1.7× less per token, which makes it the better buy when DeepSeek V4.1 Flash's lead doesn't matter for your workload.

### Which is cheaper, DeepSeek V4.1 Flash 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 V4.1 Flash lists at $0.15 and $0.60.

### Is DeepSeek V4.1 Flash or Gemma 4 31B IT better for coding?

DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 42.3 in the Noometry coding category.

### Which has the bigger context window?

DeepSeek V4.1 Flash does, with 1M tokens against 262K.

### How many benchmarks do DeepSeek V4.1 Flash and Gemma 4 31B IT share?

27 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and Gemma 4 31B IT has 35.