# DeepSeek V4.1 Flash vs Gemini 2.5 Pro

> DeepSeek V4.1 Flash is the stronger model overall, scoring 50.3 to 44.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v4-1-flash-vs-gemini-2-5-pro
- Last updated: 2026-10-10
- Shared benchmarks: 22

## Summary

- They share 22 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 6 categories and Gemini 2.5 Pro in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4.1 Flash leads 46.2 to 28.8.
- The biggest single-benchmark swing is CritPt: 14.3% for DeepSeek V4.1 Flash and 2% for Gemini 2.5 Pro.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $1.25 / $10 for Gemini 2.5 Pro.
- Gemini 2.5 Pro accepts more context: 1.05M tokens versus 1M.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek V4.1 Flash | Gemini 2.5 Pro |
|---|---|---|
| Provider | DeepSeek | Google |
| Noometry Index | 50.3 | 44.9 |
| Rank | 40 | 67 |
| Context | 1M | 1.05M |
| Input $/M | $0.15 | $1.25 |
| Output $/M | $0.60 | $10 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek V4.1 Flash: 52.8 (#29)
- Gemini 2.5 Pro: 42.8 (#86)

| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Pro |
|---|---|---|
| LMArena WebDev | 1620 | 1227 |
| SciCode | 51.9% | 42.8% |
| LMArena Coding | 1511 | 1452 |
| ALE-Bench | 1,092 | 785.52 |
| SWE-bench Verified | — | 57.6% |
| SWE-bench Verified (bash only) | — | 53.6% |
| Aider Polyglot | — | 83.1% |
| GSO | — | 3.9% |
| WeirdML | — | 54% |
| LiveBench Coding | — | 85.9% |
| CadEval | — | 64% |
| AlgoTune | — | 1.51 |

## Agentic & Tool Use

- DeepSeek V4.1 Flash: —
- Gemini 2.5 Pro: 29.1 (#81)

| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Pro |
|---|---|---|
| Terminal-Bench | — | 32.6% |
| GDPval | — | 23.3% |
| Remote Labor Index | — | 0.8% |
| TheAgentCompany | — | 30.3% |
| τ²-bench Banking | — | 13.7% |
| DeepResearch Bench | — | 42.8% |
| BALROG | — | 43.3% |
| LMArena Search | — | 1142 |
| METR Time Horizons | — | 55.4% |
| Vending-Bench 2 | — | 573.64 |

## Reasoning

- DeepSeek V4.1 Flash: 46.2 (#39)
- Gemini 2.5 Pro: 28.8 (#85)

| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Pro |
|---|---|---|
| CritPt | 14.3% | 2% |
| LMArena Hard Prompts | 1485 | 1455 |
| DTBench | 89.9% | 82.4% |
| LMCA | 47% | 34.8% |
| Epoch Capabilities Index | 155.01 | 145.32 |
| ARC-AGI-2 | — | 4.9% |
| SimpleBench | — | 62.4% |
| Kagi LLM Benchmark | — | 70.3% |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | — | 41% |
| Chess Puzzles | — | 20% |
| EnigmaEval | — | 5.6% |
| LiveBench Reasoning | — | 89.8% |
| LiveBench Data Analysis | — | 79.9% |
| Surface Evolver Bench | 46.3% | — |
| ForecastBench | — | 61.3 |
| LiveBench | — | 82.3% |

## Math

- DeepSeek V4.1 Flash: 48.4 (#52)
- Gemini 2.5 Pro: 31.9 (#191)

| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Pro |
|---|---|---|
| LMArena Math | 1493 | 1450 |
| FrontierMath (Tiers 1-3) | — | 24.6% |
| FrontierMath Tier 4 | — | 0% |
| OTIS Mock AIME 2024-2025 | — | 84.7% |
| ProofBench | 54% | — |
| Omni-MATH | — | 41.6% |
| LiveBench Math | — | 90.2% |
| MATH Level 5 | — | 95.9% |
| FrontierMath (Feb 2025 set) | — | 14.1% |
| FrontierMath Tier 4 (v1) | — | 4.2% |

## Knowledge

- DeepSeek V4.1 Flash: 41.6 (#97)
- Gemini 2.5 Pro: 55.8 (#41)

| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Pro |
|---|---|---|
| LMArena Expert | 1505 | 1452 |
| GPQA Diamond | — | 85.3% |
| Humanity's Last Exam | — | 21.6% |
| MMLU-Pro | — | 86.3% |
| Confabulations | — | 10.6% |
| Vectara Hallucination Rate | — | 7% |
| GPQA (HELM) | — | 74.9% |

## Multimodal

- DeepSeek V4.1 Flash: —
- Gemini 2.5 Pro: 44.5 (#16)

| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Pro |
|---|---|---|
| LMArena Vision | — | 1263 |
| GeoBench | — | 86% |
| VPCT | — | 48% |
| LMArena Document | — | 1421 |
| SpatialViz-Bench | — | 44.7% |

## Multilingual

- DeepSeek V4.1 Flash: 55.5 (#29)
- Gemini 2.5 Pro: 55.5 (#30)

| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Pro |
|---|---|---|
| LMArena Non-English | 1452 | 1452 |
| LMArena Chinese | 1506 | 1507 |
| LMArena French | 1448 | 1473 |
| LMArena Russian | 1473 | 1460 |
| LMArena Spanish | 1499 | 1473 |
| LMArena German | — | 1487 |
| LMArena Japanese | — | 1462 |
| LMArena Korean | — | 1435 |

## Instruction Following

- DeepSeek V4.1 Flash: 77.6 (#24)
- Gemini 2.5 Pro: 75.2 (#71)

| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Pro |
|---|---|---|
| LMArena Instruction Following | 1475 | 1436 |
| LiveBench Instruction Following | — | 80.6% |
| IFEval | — | 84% |

## Long Context

- DeepSeek V4.1 Flash: 45.2 (#46)
- Gemini 2.5 Pro: 59.7 (#5)

| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Pro |
|---|---|---|
| LMArena Longer Query | 1476 | 1449 |
| Fiction.LiveBench | — | 91.7% |

## Writing & Preference

- DeepSeek V4.1 Flash: 65.8 (#43)
- Gemini 2.5 Pro: 63.9 (#60)

| Benchmark | DeepSeek V4.1 Flash | Gemini 2.5 Pro |
|---|---|---|
| LMArena Text | 1465 | 1458 |
| LMArena Creative Writing | 1436 | 1453 |
| EQ-Bench Creative Writing | 1540 | 1421 |
| LMArena Multi-Turn | 1460 | 1453 |
| Short-Story Creative Writing | — | 83.8% |
| WildBench | — | 85.7% |
| LiveBench Language | — | 67.8% |

## FAQ

### Is DeepSeek V4.1 Flash better than Gemini 2.5 Pro?

DeepSeek V4.1 Flash is the stronger model overall, scoring 50.3 to 44.9 on the Noometry Index.

### Which is cheaper, DeepSeek V4.1 Flash or Gemini 2.5 Pro?

DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Gemini 2.5 Pro lists at $1.25 and $10.

### Is DeepSeek V4.1 Flash or Gemini 2.5 Pro better for coding?

DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.8 versus 42.8 in the Noometry coding category.

### Which has the bigger context window?

Gemini 2.5 Pro does, with 1.05M tokens against 1M.

### How many benchmarks do DeepSeek V4.1 Flash and Gemini 2.5 Pro share?

22 benchmarks have published results for both models. DeepSeek V4.1 Flash has 25 scored results on Noometry and Gemini 2.5 Pro has 78.