# DeepSeek V4.1 Flash vs o4-mini

> DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 41.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v4-1-flash-vs-o4-mini
- Last updated: 2026-10-11
- Shared benchmarks: 28

## Summary

- They share 28 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 7 categories and o4-mini in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4.1 Flash leads 66.7 to 40.8.
- The biggest single-benchmark swing is Mystery Game Puzzles: 43% for DeepSeek V4.1 Flash and 5% for o4-mini.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $1.10 / $4.40 for o4-mini.
- DeepSeek V4.1 Flash accepts more context: 1M tokens versus 200K.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek V4.1 Flash | o4-mini |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 52.8 | 41.6 |
| Rank | 38 | 132 |
| Context | 1M | 200K |
| Input $/M | $0.15 | $1.10 |
| Output $/M | $0.60 | $4.40 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek V4.1 Flash: 52.9 (#32)
- o4-mini: 40.9 (#127)

| Benchmark | DeepSeek V4.1 Flash | o4-mini |
|---|---|---|
| LMArena Coding | 1506 | 1368 |
| ALE-Bench | 1,092 | 826.17 |
| SWE-bench Verified (bash only) | — | 45% |
| Aider Polyglot | — | 72% |
| LMArena WebDev | 1619 | — |
| SciCode | 51.9% | — |
| GSO | — | 3.6% |
| WeirdML | — | 52.6% |
| CadEval | — | 62% |
| AlgoTune | — | 1.72 |

## Agentic & Tool Use

- DeepSeek V4.1 Flash: 31.2 (#69)
- o4-mini: 32.6 (#61)

| Benchmark | DeepSeek V4.1 Flash | o4-mini |
|---|---|---|
| APEX-Agents | 39.5% | — |
| Berkeley Function Calling Leaderboard | — | 53.2% |
| GDPval | — | 25.3% |
| GDP.pdf | 19.8% | — |
| METR Time Horizons | — | 63.9% |

## Reasoning

- DeepSeek V4.1 Flash: 50.2 (#36)
- o4-mini: 24.6 (#162)

| Benchmark | DeepSeek V4.1 Flash | o4-mini |
|---|---|---|
| CritPt | 14.3% | 0.6% |
| LMArena Hard Prompts | 1483 | 1351 |
| Mystery Game Puzzles | 43% | 5% |
| DTBench | 89.9% | 77.6% |
| LMCA | 47% | 26.5% |
| Epoch Capabilities Index | 154.9 | 145.64 |
| ARC-AGI-2 | — | 6.1% |
| SimpleBench | — | 38.7% |
| Kagi LLM Benchmark | — | 67.6% |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | — | 58.7% |
| Chess Puzzles | — | 26% |
| EnigmaEval | — | 9.2% |
| Surface Evolver Bench | 46.3% | — |
| ForecastBench | — | 61.8 |

## Math

- DeepSeek V4.1 Flash: 66.7 (#25)
- o4-mini: 40.8 (#89)

| Benchmark | DeepSeek V4.1 Flash | o4-mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 67.4% | 36.1% |
| FrontierMath Tier 4 | 26.8% | 4.9% |
| OTIS Mock AIME 2024-2025 | 98.3% | 81.7% |
| LMArena Math | 1477 | 1389 |
| ProofBench | 54% | — |
| Omni-MATH | — | 72% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 24.8% |
| FrontierMath Tier 4 (v1) | — | 6.3% |

## Knowledge

- DeepSeek V4.1 Flash: 57.9 (#38)
- o4-mini: 43.6 (#91)

| Benchmark | DeepSeek V4.1 Flash | o4-mini |
|---|---|---|
| GPQA Diamond | 89.8% | 79.6% |
| LMArena Expert | 1506 | 1343 |
| Humanity's Last Exam | — | 18.1% |
| SimpleQA Verified | — | 19.6% |
| MMLU-Pro | — | 82% |
| Confabulations | — | 15.8% |
| Vectara Hallucination Rate | — | 18.6% |
| GPQA (HELM) | — | 73.5% |

## Multimodal

- DeepSeek V4.1 Flash: 39.1 (#61)
- o4-mini: 40.2 (#49)

| Benchmark | DeepSeek V4.1 Flash | o4-mini |
|---|---|---|
| LMArena Vision | 1277 | 1194 |
| GeoBench | — | 64% |
| VPCT | — | 57.5% |
| Furniture Assembly | 34.2% | — |

## Multilingual

- DeepSeek V4.1 Flash: 55.0 (#35)
- o4-mini: 47.0 (#154)

| Benchmark | DeepSeek V4.1 Flash | o4-mini |
|---|---|---|
| LMArena Non-English | 1448 | 1337 |
| LMArena Chinese | 1497 | 1354 |
| LMArena French | 1452 | 1364 |
| LMArena German | 1484 | 1336 |
| LMArena Japanese | 1412 | 1308 |
| LMArena Korean | 1452 | 1312 |
| LMArena Russian | 1471 | 1334 |
| LMArena Spanish | 1459 | 1347 |

## Instruction Following

- DeepSeek V4.1 Flash: 77.3 (#26)
- o4-mini: 75.2 (#68)

| Benchmark | DeepSeek V4.1 Flash | o4-mini |
|---|---|---|
| LMArena Instruction Following | 1474 | 1321 |
| IFEval | — | 92.8% |

## Long Context

- DeepSeek V4.1 Flash: 45.2 (#47)
- o4-mini: 45.5 (#33)

| Benchmark | DeepSeek V4.1 Flash | o4-mini |
|---|---|---|
| LMArena Longer Query | 1475 | 1315 |
| Fiction.LiveBench | — | 77.8% |

## Writing & Preference

- DeepSeek V4.1 Flash: 65.4 (#48)
- o4-mini: 54.0 (#152)

| Benchmark | DeepSeek V4.1 Flash | o4-mini |
|---|---|---|
| LMArena Text | 1462 | 1353 |
| LMArena Creative Writing | 1435 | 1294 |
| LMArena Multi-Turn | 1457 | 1350 |
| Short-Story Creative Writing | — | 75% |
| EQ-Bench Creative Writing | 1540 | — |
| WildBench | — | 85.4% |

## FAQ

### Is DeepSeek V4.1 Flash better than o4-mini?

DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 41.6 on the Noometry Index.

### Which is cheaper, DeepSeek V4.1 Flash or o4-mini?

DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; o4-mini lists at $1.10 and $4.40.

### Is DeepSeek V4.1 Flash or o4-mini better for coding?

DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 40.9 in the Noometry coding category.

### Which has the bigger context window?

DeepSeek V4.1 Flash does, with 1M tokens against 200K.

### How many benchmarks do DeepSeek V4.1 Flash and o4-mini share?

28 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and o4-mini has 60.