# DeepSeek-V3.1 vs Gemini 3.6 Flash

> Gemini 3.6 Flash is the stronger model overall, scoring 54.1 to 42.2 on the Noometry Index. DeepSeek-V3.1 costs 3.5× less per token, which makes it the better buy when Gemini 3.6 Flash's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/deepseek-v3-1-vs-gemini-3-6-flash
- Last updated: 2026-10-10
- Shared benchmarks: 22

## Summary

- They share 22 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 0 categories and Gemini 3.6 Flash in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.6 Flash leads 58.9 to 27.3.
- The biggest single-benchmark swing is LMCA: 24.3% for DeepSeek-V3.1 and 44.9% for Gemini 3.6 Flash.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $0.75 / $3.75 for Gemini 3.6 Flash.
- Gemini 3.6 Flash accepts more context: 1.05M tokens versus 164K.
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek-V3.1 | Gemini 3.6 Flash |
|---|---|---|
| Provider | DeepSeek | Google |
| Noometry Index | 42.2 | 54.1 |
| Rank | 106 | 30 |
| Context | 164K | 1.05M |
| Input $/M | $0.25 | $0.75 |
| Output $/M | $0.95 | $3.75 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek-V3.1: 39.8 (#140)
- Gemini 3.6 Flash: 49.6 (#45)

| Benchmark | DeepSeek-V3.1 | Gemini 3.6 Flash |
|---|---|---|
| WeirdML | 38.4% | 56.1% |
| LMArena Coding | 1417 | 1492 |
| DeepSWE | — | 46.7% |
| FrontierCode | — | 34.4% |
| LMArena WebDev | — | 1536 |
| SciCode | — | 52.7% |
| ALE-Bench | — | 715.52 |

## Agentic & Tool Use

- DeepSeek-V3.1: —
- Gemini 3.6 Flash: 32.4 (#56)

| Benchmark | DeepSeek-V3.1 | Gemini 3.6 Flash |
|---|---|---|
| APEX-Agents | — | 46.9% |
| GDP.pdf | — | 14% |

## Reasoning

- DeepSeek-V3.1: 27.3 (#109)
- Gemini 3.6 Flash: 58.9 (#20)

| Benchmark | DeepSeek-V3.1 | Gemini 3.6 Flash |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1486 |
| DTBench | 82.7% | 95.5% |
| LMCA | 24.3% | 44.9% |
| Epoch Capabilities Index | 139.92 | 154.36 |
| ARC-AGI-2 | — | 60.4% |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| NYT Connections (extended) | — | 89% |
| ARC-AGI-1 | — | 91.2% |
| CritPt | — | 10.6% |
| Chess Puzzles | — | 43% |
| Mystery Game Puzzles | — | 30% |
| ForecastBench | 58 | — |

## Math

- DeepSeek-V3.1: 36.1 (#125)
- Gemini 3.6 Flash: 57.0 (#33)

| Benchmark | DeepSeek-V3.1 | Gemini 3.6 Flash |
|---|---|---|
| LMArena Math | 1420 | 1504 |
| FrontierMath (Tiers 1-3) | — | 58.9% |
| FrontierMath Tier 4 | — | 22% |
| MathArena Final-Answer Competitions | — | 70.8% |
| OTIS Mock AIME 2024-2025 | — | 94.2% |
| ProofBench | — | 36% |

## Knowledge

- DeepSeek-V3.1: 43.0 (#86)
- Gemini 3.6 Flash: 67.8 (#7)

| Benchmark | DeepSeek-V3.1 | Gemini 3.6 Flash |
|---|---|---|
| LMArena Expert | 1403 | 1489 |
| GPQA Diamond | — | 94.1% |
| SimpleQA Verified | — | 66.2% |
| Vectara Hallucination Rate | 5.5% | — |

## Multimodal

- DeepSeek-V3.1: —
- Gemini 3.6 Flash: 38.7 (#55)

| Benchmark | DeepSeek-V3.1 | Gemini 3.6 Flash |
|---|---|---|
| LMArena Vision | — | 1297 |
| Blueprint-Bench 2 | — | 31.2% |
| Furniture Assembly | — | 23.3% |
| LMArena Document | — | 1456 |

## Multilingual

- DeepSeek-V3.1: 51.8 (#100)
- Gemini 3.6 Flash: 56.8 (#17)

| Benchmark | DeepSeek-V3.1 | Gemini 3.6 Flash |
|---|---|---|
| LMArena Non-English | 1400 | 1470 |
| LMArena Chinese | 1470 | 1530 |
| LMArena French | 1447 | 1501 |
| LMArena German | 1412 | 1478 |
| LMArena Japanese | 1374 | 1477 |
| LMArena Korean | 1337 | 1435 |
| LMArena Russian | 1405 | 1490 |
| LMArena Spanish | 1429 | 1477 |

## Instruction Following

- DeepSeek-V3.1: 74.1 (#105)
- Gemini 3.6 Flash: 77.2 (#27)

| Benchmark | DeepSeek-V3.1 | Gemini 3.6 Flash |
|---|---|---|
| LMArena Instruction Following | 1400 | 1467 |

## Long Context

- DeepSeek-V3.1: 36.3 (#227)
- Gemini 3.6 Flash: 45.1 (#47)

| Benchmark | DeepSeek-V3.1 | Gemini 3.6 Flash |
|---|---|---|
| LMArena Longer Query | 1422 | 1475 |
| Fiction.LiveBench | 52.8% | — |

## Writing & Preference

- DeepSeek-V3.1: 60.5 (#93)
- Gemini 3.6 Flash: 68.3 (#26)

| Benchmark | DeepSeek-V3.1 | Gemini 3.6 Flash |
|---|---|---|
| LMArena Text | 1420 | 1478 |
| LMArena Creative Writing | 1401 | 1464 |
| EQ-Bench Creative Writing | 1436 | 1604 |
| LMArena Multi-Turn | 1408 | 1481 |

## FAQ

### Is DeepSeek-V3.1 better than Gemini 3.6 Flash?

Gemini 3.6 Flash is the stronger model overall, scoring 54.1 to 42.2 on the Noometry Index. DeepSeek-V3.1 costs 3.5× less per token, which makes it the better buy when Gemini 3.6 Flash's lead doesn't matter for your workload.

### Which is cheaper, DeepSeek-V3.1 or Gemini 3.6 Flash?

DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; Gemini 3.6 Flash lists at $0.75 and $3.75.

### Is DeepSeek-V3.1 or Gemini 3.6 Flash better for coding?

Gemini 3.6 Flash scores higher on coding benchmarks: 49.6 versus 39.8 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.6 Flash does, with 1.05M tokens against 164K.

### How many benchmarks do DeepSeek-V3.1 and Gemini 3.6 Flash share?

22 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Gemini 3.6 Flash has 46.