# DeepSeek V4 Flash vs Gemini 3.6 Flash

> DeepSeek V4 Flash and Gemini 3.6 Flash score almost the same on the Noometry Index (53.6 vs 54.1), so choose on price, context window or the category you care about most.

- Canonical page: https://noometry.com/compare/deepseek-v4-flash-vs-gemini-3-6-flash
- Last updated: 2026-10-10
- Shared benchmarks: 39

## Summary

- They share 39 benchmarks with published results for both. DeepSeek V4 Flash scores higher in 1 category and Gemini 3.6 Flash in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 3.6 Flash leads 67.8 to 55.4.
- The biggest single-benchmark swing is SimpleQA Verified: 33.6% for DeepSeek V4 Flash and 66.2% for Gemini 3.6 Flash.
- DeepSeek V4 Flash is cheaper at $0.15 / $0.60 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 1M.
- DeepSeek V4 Flash has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek V4 Flash | Gemini 3.6 Flash |
|---|---|---|
| Provider | DeepSeek | Google |
| Noometry Index | 53.6 | 54.1 |
| Rank | 35 | 33 |
| Context | 1M | 1.05M |
| Input $/M | $0.15 | $0.75 |
| Output $/M | $0.60 | $3.75 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek V4 Flash: 47.9 (#59)
- Gemini 3.6 Flash: 50.0 (#48)

| Benchmark | DeepSeek V4 Flash | Gemini 3.6 Flash |
|---|---|---|
| FrontierCode | 18.8% | 34.4% |
| LMArena WebDev | 1582 | 1538 |
| SciCode | 49.9% | 52.7% |
| WeirdML | 63% | 56.1% |
| LMArena Coding | 1457 | 1491 |
| ALE-Bench | 1,306 | 715.52 |
| DeepSWE | — | 46.7% |

## Agentic & Tool Use

- DeepSeek V4 Flash: —
- Gemini 3.6 Flash: 32.3 (#65)

| Benchmark | DeepSeek V4 Flash | Gemini 3.6 Flash |
|---|---|---|
| APEX-Agents | — | 46.9% |
| GDP.pdf | — | 14% |

## Reasoning

- DeepSeek V4 Flash: 53.7 (#30)
- Gemini 3.6 Flash: 58.8 (#22)

| Benchmark | DeepSeek V4 Flash | Gemini 3.6 Flash |
|---|---|---|
| ARC-AGI-2 | 61.4% | 60.4% |
| NYT Connections (extended) | 89.6% | 89% |
| ARC-AGI-1 | 89% | 91.2% |
| CritPt | 16.6% | 10.6% |
| Chess Puzzles | 33% | 43% |
| LMArena Hard Prompts | 1444 | 1485 |
| Mystery Game Puzzles | 34% | 30% |
| DTBench | 90.9% | 95.5% |
| LMCA | 41.7% | 44.9% |
| Epoch Capabilities Index | 154.49 | 154.25 |
| SimpleBench | 61.1% | — |
| Kagi LLM Benchmark | 52.2% | — |

## Math

- DeepSeek V4 Flash: 60.3 (#37)
- Gemini 3.6 Flash: 57.3 (#40)

| Benchmark | DeepSeek V4 Flash | Gemini 3.6 Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 57.5% | 58.9% |
| FrontierMath Tier 4 | 24.4% | 22% |
| MathArena Final-Answer Competitions | 76.5% | 70.8% |
| OTIS Mock AIME 2024-2025 | 94.4% | 94.2% |
| ProofBench | 56% | 36% |
| LMArena Math | 1427 | 1505 |

## Knowledge

- DeepSeek V4 Flash: 55.4 (#48)
- Gemini 3.6 Flash: 67.8 (#8)

| Benchmark | DeepSeek V4 Flash | Gemini 3.6 Flash |
|---|---|---|
| GPQA Diamond | 91% | 94.1% |
| SimpleQA Verified | 33.6% | 66.2% |
| LMArena Expert | 1441 | 1488 |

## Multimodal

- DeepSeek V4 Flash: —
- Gemini 3.6 Flash: 38.5 (#64)

| Benchmark | DeepSeek V4 Flash | Gemini 3.6 Flash |
|---|---|---|
| LMArena Vision | — | 1298 |
| Blueprint-Bench 2 | — | 31.2% |
| Furniture Assembly | — | 23.3% |
| LMArena Document | — | 1456 |

## Multilingual

- DeepSeek V4 Flash: 53.0 (#72)
- Gemini 3.6 Flash: 56.5 (#19)

| Benchmark | DeepSeek V4 Flash | Gemini 3.6 Flash |
|---|---|---|
| LMArena Non-English | 1420 | 1469 |
| LMArena Chinese | 1468 | 1531 |
| LMArena French | 1439 | 1504 |
| LMArena German | 1418 | 1478 |
| LMArena Japanese | 1406 | 1476 |
| LMArena Korean | 1384 | 1431 |
| LMArena Russian | 1428 | 1487 |
| LMArena Spanish | 1436 | 1475 |

## Instruction Following

- DeepSeek V4 Flash: 74.9 (#81)
- Gemini 3.6 Flash: 77.0 (#33)

| Benchmark | DeepSeek V4 Flash | Gemini 3.6 Flash |
|---|---|---|
| LMArena Instruction Following | 1421 | 1466 |

## Long Context

- DeepSeek V4 Flash: 43.8 (#85)
- Gemini 3.6 Flash: 45.1 (#50)

| Benchmark | DeepSeek V4 Flash | Gemini 3.6 Flash |
|---|---|---|
| LMArena Longer Query | 1434 | 1474 |

## Writing & Preference

- DeepSeek V4 Flash: 63.8 (#61)
- Gemini 3.6 Flash: 68.2 (#27)

| Benchmark | DeepSeek V4 Flash | Gemini 3.6 Flash |
|---|---|---|
| LMArena Text | 1432 | 1479 |
| LMArena Creative Writing | 1403 | 1465 |
| EQ-Bench Creative Writing | 1559 | 1604 |
| LMArena Multi-Turn | 1449 | 1481 |

## FAQ

### Is DeepSeek V4 Flash better than Gemini 3.6 Flash?

DeepSeek V4 Flash and Gemini 3.6 Flash score almost the same on the Noometry Index (53.6 vs 54.1), so choose on price, context window or the category you care about most.

### Which is cheaper, DeepSeek V4 Flash or Gemini 3.6 Flash?

DeepSeek V4 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Gemini 3.6 Flash lists at $0.75 and $3.75.

### Is DeepSeek V4 Flash or Gemini 3.6 Flash better for coding?

Gemini 3.6 Flash scores higher on coding benchmarks: 50.0 versus 47.9 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.6 Flash does, with 1.05M tokens against 1M.

### How many benchmarks do DeepSeek V4 Flash and Gemini 3.6 Flash share?

39 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and Gemini 3.6 Flash has 46.