# DeepSeek V4.1 Flash vs GPT-4.5

> DeepSeek V4.1 Flash is the stronger model overall, scoring 50.3 to 36.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v4-1-flash-vs-gpt-4-5
- Last updated: 2026-10-10
- Shared benchmarks: 15

## Summary

- They share 15 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 8 categories and GPT-4.5 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4.1 Flash leads 46.2 to 14.1.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek V4.1 Flash | GPT-4.5 |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 50.3 | 36.9 |
| Rank | 40 | 200 |
| Context | 1M | — |
| Input $/M | $0.15 | — |
| Output $/M | $0.60 | — |
| Weights | Open | Proprietary |

## Coding

- DeepSeek V4.1 Flash: 52.8 (#29)
- GPT-4.5: 41.8 (#102)

| Benchmark | DeepSeek V4.1 Flash | GPT-4.5 |
|---|---|---|
| LMArena Coding | 1511 | 1397 |
| Aider Polyglot | — | 44.9% |
| LMArena WebDev | 1620 | — |
| SciCode | 51.9% | — |
| WeirdML | — | 39.4% |
| LiveBench Coding | — | 75.2% |
| ALE-Bench | 1,092 | — |

## Agentic & Tool Use

- DeepSeek V4.1 Flash: —
- GPT-4.5: 27.6 (#92)

| Benchmark | DeepSeek V4.1 Flash | GPT-4.5 |
|---|---|---|
| Cybench | — | 17.5% |

## Reasoning

- DeepSeek V4.1 Flash: 46.2 (#39)
- GPT-4.5: 14.1 (#320)

| Benchmark | DeepSeek V4.1 Flash | GPT-4.5 |
|---|---|---|
| LMArena Hard Prompts | 1485 | 1404 |
| Epoch Capabilities Index | 155.01 | 136.74 |
| ARC-AGI-2 | — | 0.8% |
| SimpleBench | — | 34.5% |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | — | 10.3% |
| CritPt | 14.3% | — |
| EnigmaEval | — | 3.2% |
| LiveBench Reasoning | — | 71.1% |
| DTBench | 89.9% | — |
| LiveBench Data Analysis | — | 64.3% |
| LMCA | 47% | — |
| Surface Evolver Bench | 46.3% | — |
| ForecastBench | — | 61.7 |
| LiveBench | — | 69% |

## Math

- DeepSeek V4.1 Flash: 48.4 (#52)
- GPT-4.5: 31.0 (#200)

| Benchmark | DeepSeek V4.1 Flash | GPT-4.5 |
|---|---|---|
| LMArena Math | 1493 | 1412 |
| OTIS Mock AIME 2024-2025 | — | 37.8% |
| ProofBench | 54% | — |
| LiveBench Math | — | 69.3% |
| MATH Level 5 | — | 78.6% |

## Knowledge

- DeepSeek V4.1 Flash: 41.6 (#97)
- GPT-4.5: 32.4 (#205)

| Benchmark | DeepSeek V4.1 Flash | GPT-4.5 |
|---|---|---|
| LMArena Expert | 1505 | 1394 |
| GPQA Diamond | — | 68.7% |
| Humanity's Last Exam | — | 5.4% |
| Confabulations | — | 13.6% |

## Multimodal

- DeepSeek V4.1 Flash: —
- GPT-4.5: 37.0 (#68)

| Benchmark | DeepSeek V4.1 Flash | GPT-4.5 |
|---|---|---|
| LMArena Vision | — | 1195 |
| VPCT | — | 45% |

## Multilingual

- DeepSeek V4.1 Flash: 55.5 (#29)
- GPT-4.5: 52.7 (#78)

| Benchmark | DeepSeek V4.1 Flash | GPT-4.5 |
|---|---|---|
| LMArena Non-English | 1452 | 1413 |
| LMArena Chinese | 1506 | 1421 |
| LMArena French | 1448 | 1417 |
| LMArena Russian | 1473 | 1419 |
| LMArena German | — | 1457 |
| LMArena Japanese | — | 1416 |
| LMArena Korean | — | 1392 |
| LMArena Spanish | 1499 | — |

## Instruction Following

- DeepSeek V4.1 Flash: 77.6 (#24)
- GPT-4.5: 72.8 (#129)

| Benchmark | DeepSeek V4.1 Flash | GPT-4.5 |
|---|---|---|
| LMArena Instruction Following | 1475 | 1404 |
| LiveBench Instruction Following | — | 72.3% |

## Long Context

- DeepSeek V4.1 Flash: 45.2 (#46)
- GPT-4.5: 40.4 (#150)

| Benchmark | DeepSeek V4.1 Flash | GPT-4.5 |
|---|---|---|
| LMArena Longer Query | 1476 | 1406 |
| Fiction.LiveBench | — | 63.9% |

## Writing & Preference

- DeepSeek V4.1 Flash: 65.8 (#43)
- GPT-4.5: 57.1 (#129)

| Benchmark | DeepSeek V4.1 Flash | GPT-4.5 |
|---|---|---|
| LMArena Text | 1465 | 1417 |
| LMArena Creative Writing | 1436 | 1394 |
| EQ-Bench Creative Writing | 1540 | 1258 |
| LMArena Multi-Turn | 1460 | 1444 |
| Short-Story Creative Writing | — | 75.6% |
| LiveBench Language | — | 61.5% |

## FAQ

### Is DeepSeek V4.1 Flash better than GPT-4.5?

DeepSeek V4.1 Flash is the stronger model overall, scoring 50.3 to 36.9 on the Noometry Index.

### Is DeepSeek V4.1 Flash or GPT-4.5 better for coding?

DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.8 versus 41.8 in the Noometry coding category.

### How many benchmarks do DeepSeek V4.1 Flash and GPT-4.5 share?

15 benchmarks have published results for both models. DeepSeek V4.1 Flash has 25 scored results on Noometry and GPT-4.5 has 42.