# DeepSeek V4.1 Flash vs GPT-5 Mini

> DeepSeek V4.1 Flash is the stronger model overall, scoring 50.3 to 41.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v4-1-flash-vs-gpt-5-mini
- 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 7 categories and GPT-5 Mini in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4.1 Flash leads 46.2 to 24.1.
- The biggest single-benchmark swing is ProofBench: 54% for DeepSeek V4.1 Flash and 9% for GPT-5 Mini.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.25 / $2 for GPT-5 Mini.
- DeepSeek V4.1 Flash accepts more context: 1M tokens versus 400K.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek V4.1 Flash | GPT-5 Mini |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 50.3 | 41.7 |
| Rank | 40 | 115 |
| Context | 1M | 400K |
| Input $/M | $0.15 | $0.25 |
| Output $/M | $0.60 | $2 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek V4.1 Flash: 52.8 (#29)
- GPT-5 Mini: 40.0 (#134)

| Benchmark | DeepSeek V4.1 Flash | GPT-5 Mini |
|---|---|---|
| SciCode | 51.9% | 39.2% |
| LMArena Coding | 1511 | 1406 |
| ALE-Bench | 1,092 | 799.77 |
| SWE-bench Verified | — | 64.7% |
| SWE-bench Verified (bash only) | — | 59.8% |
| LMArena WebDev | 1620 | — |
| SWE-bench Multilingual | — | 39.7% |
| WeirdML | — | 52.7% |
| AlgoTune | — | 1.38 |

## Agentic & Tool Use

- DeepSeek V4.1 Flash: —
- GPT-5 Mini: 31.0 (#64)

| Benchmark | DeepSeek V4.1 Flash | GPT-5 Mini |
|---|---|---|
| Terminal-Bench | — | 34.8% |
| Berkeley Function Calling Leaderboard | — | 55.5% |
| Vending-Bench 2 | — | -31.18 |

## Reasoning

- DeepSeek V4.1 Flash: 46.2 (#39)
- GPT-5 Mini: 24.1 (#153)

| Benchmark | DeepSeek V4.1 Flash | GPT-5 Mini |
|---|---|---|
| CritPt | 14.3% | 0% |
| LMArena Hard Prompts | 1485 | 1380 |
| DTBench | 89.9% | 80.5% |
| LMCA | 47% | 34.2% |
| Epoch Capabilities Index | 155.01 | 145.51 |
| ARC-AGI-2 | — | 4.4% |
| Kagi LLM Benchmark | — | 70.3% |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | — | 54.3% |
| Chess Puzzles | — | 30% |
| EnigmaEval | — | 8.2% |
| Mystery Game Puzzles | — | 10% |
| Surface Evolver Bench | 46.3% | — |
| ForecastBench | — | 61 |

## Math

- DeepSeek V4.1 Flash: 48.4 (#52)
- GPT-5 Mini: 46.3 (#59)

| Benchmark | DeepSeek V4.1 Flash | GPT-5 Mini |
|---|---|---|
| ProofBench | 54% | 9% |
| LMArena Math | 1493 | 1377 |
| FrontierMath (Tiers 1-3) | — | 46.7% |
| FrontierMath Tier 4 | — | 12.2% |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| Omni-MATH | — | 72.2% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 27.2% |
| FrontierMath Tier 4 (v1) | — | 6.3% |

## Knowledge

- DeepSeek V4.1 Flash: 41.6 (#97)
- GPT-5 Mini: 45.5 (#80)

| Benchmark | DeepSeek V4.1 Flash | GPT-5 Mini |
|---|---|---|
| LMArena Expert | 1505 | 1377 |
| GPQA Diamond | — | 75% |
| Humanity's Last Exam | — | 19.4% |
| SimpleQA Verified | — | 21.6% |
| MMLU-Pro | — | 83.5% |
| Confabulations | — | 13.3% |
| Vectara Hallucination Rate | — | 12.9% |
| GPQA (HELM) | — | 75.6% |

## Multimodal

- DeepSeek V4.1 Flash: —
- GPT-5 Mini: 35.0 (#80)

| Benchmark | DeepSeek V4.1 Flash | GPT-5 Mini |
|---|---|---|
| LMArena Vision | — | 1202 |
| VPCT | — | 40.2% |

## Multilingual

- DeepSeek V4.1 Flash: 55.5 (#29)
- GPT-5 Mini: 49.1 (#131)

| Benchmark | DeepSeek V4.1 Flash | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1452 | 1363 |
| LMArena Chinese | 1506 | 1387 |
| LMArena French | 1448 | 1385 |
| LMArena Russian | 1473 | 1362 |
| LMArena Spanish | 1499 | 1355 |
| LMArena German | — | 1366 |
| LMArena Japanese | — | 1342 |
| LMArena Korean | — | 1306 |

## Instruction Following

- DeepSeek V4.1 Flash: 77.6 (#24)
- GPT-5 Mini: 76.4 (#43)

| Benchmark | DeepSeek V4.1 Flash | GPT-5 Mini |
|---|---|---|
| LMArena Instruction Following | 1475 | 1357 |
| IFEval | — | 92.7% |

## Long Context

- DeepSeek V4.1 Flash: 45.2 (#46)
- GPT-5 Mini: 41.9 (#127)

| Benchmark | DeepSeek V4.1 Flash | GPT-5 Mini |
|---|---|---|
| LMArena Longer Query | 1476 | 1355 |
| Fiction.LiveBench | — | 69.4% |

## Writing & Preference

- DeepSeek V4.1 Flash: 65.8 (#43)
- GPT-5 Mini: 55.4 (#143)

| Benchmark | DeepSeek V4.1 Flash | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1465 | 1373 |
| LMArena Creative Writing | 1436 | 1325 |
| EQ-Bench Creative Writing | 1540 | 1313 |
| LMArena Multi-Turn | 1460 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
| WildBench | — | 85.5% |

## FAQ

### Is DeepSeek V4.1 Flash better than GPT-5 Mini?

DeepSeek V4.1 Flash is the stronger model overall, scoring 50.3 to 41.7 on the Noometry Index.

### Which is cheaper, DeepSeek V4.1 Flash or GPT-5 Mini?

DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GPT-5 Mini lists at $0.25 and $2.

### Is DeepSeek V4.1 Flash or GPT-5 Mini better for coding?

DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.8 versus 40.0 in the Noometry coding category.

### Which has the bigger context window?

DeepSeek V4.1 Flash does, with 1M tokens against 400K.

### How many benchmarks do DeepSeek V4.1 Flash and GPT-5 Mini share?

22 benchmarks have published results for both models. DeepSeek V4.1 Flash has 25 scored results on Noometry and GPT-5 Mini has 60.