# DeepSeek V4 Pro vs GPT-4.1

> DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 35.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v4-pro-vs-gpt-4-1
- Last updated: 2026-10-10
- Shared benchmarks: 34

## Summary

- They share 34 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 8 categories and GPT-4.1 in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Pro leads 56.5 to 11.7.
- The biggest single-benchmark swing is ARC-AGI-1: 90.5% for DeepSeek V4 Pro and 5.5% for GPT-4.1.
- DeepSeek V4 Pro is cheaper at $0.66 / $1.98 per million input/output tokens, against $2 / $8 for GPT-4.1.
- GPT-4.1 accepts more context: 1.05M tokens versus 1M.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek V4 Pro | GPT-4.1 |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 54.3 | 35.9 |
| Rank | 31 | 219 |
| Context | 1M | 1.05M |
| Input $/M | $0.66 | $2 |
| Output $/M | $1.98 | $8 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek V4 Pro: 52.4 (#34)
- GPT-4.1: 34.4 (#238)

| Benchmark | DeepSeek V4 Pro | GPT-4.1 |
|---|---|---|
| SWE-bench Verified | 77.6% | 48.5% |
| WeirdML | 66.2% | 39% |
| LMArena Coding | 1470 | 1391 |
| ALE-Bench | 1,403 | 558.1 |
| FrontierCode | 28.6% | — |
| SWE-bench Verified (bash only) | — | 39.6% |
| Aider Polyglot | — | 52.4% |
| LMArena WebDev | 1582 | — |
| SciCode | 51% | — |
| CadEval | — | 42% |

## Agentic & Tool Use

- DeepSeek V4 Pro: 32.8 (#58)
- GPT-4.1: 34.7 (#43)

| Benchmark | DeepSeek V4 Pro | GPT-4.1 |
|---|---|---|
| APEX-Agents | 47.3% | — |
| Berkeley Function Calling Leaderboard | — | 54% |
| Vending-Bench 2 | 3,285 | — |

## Reasoning

- DeepSeek V4 Pro: 56.5 (#24)
- GPT-4.1: 11.7 (#339)

| Benchmark | DeepSeek V4 Pro | GPT-4.1 |
|---|---|---|
| ARC-AGI-2 | 61.3% | 0.4% |
| Kagi LLM Benchmark | 53.5% | 52.3% |
| ARC-AGI-1 | 90.5% | 5.5% |
| Chess Puzzles | 47% | 6% |
| LMArena Hard Prompts | 1461 | 1384 |
| DTBench | 93.9% | 68.3% |
| LMCA | 45.5% | 25.6% |
| Epoch Capabilities Index | 155.31 | 136.78 |
| ForecastBench | 56.1 | 61.5 |
| SimpleBench | — | 27% |
| NYT Connections (extended) | 91.3% | — |
| CritPt | 18% | — |
| EnigmaEval | — | 2.2% |
| Mystery Game Puzzles | 43% | — |
| Surface Evolver Bench | 40% | — |

## Math

- DeepSeek V4 Pro: 64.8 (#30)
- GPT-4.1: 22.3 (#280)

| Benchmark | DeepSeek V4 Pro | GPT-4.1 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 64.6% | 6% |
| OTIS Mock AIME 2024-2025 | 98.6% | 38.3% |
| LMArena Math | 1455 | 1370 |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 76.6% | — |
| ProofBench | 50% | — |
| Omni-MATH | — | 47.1% |
| MATH Level 5 | — | 83% |
| FrontierMath (Feb 2025 set) | — | 5.5% |
| FrontierMath Tier 4 (v1) | — | 0% |

## Knowledge

- DeepSeek V4 Pro: 59.5 (#31)
- GPT-4.1: 37.1 (#160)

| Benchmark | DeepSeek V4 Pro | GPT-4.1 |
|---|---|---|
| GPQA Diamond | 91.7% | 66.9% |
| SimpleQA Verified | 52.9% | 31.1% |
| Vectara Hallucination Rate | 8.6% | 5.6% |
| LMArena Expert | 1464 | 1364 |
| Humanity's Last Exam | — | 5.4% |
| MMLU-Pro | — | 81.1% |
| GPQA (HELM) | — | 65.9% |

## Multimodal

- DeepSeek V4 Pro: —
- GPT-4.1: 38.2 (#67)

| Benchmark | DeepSeek V4 Pro | GPT-4.1 |
|---|---|---|
| LMArena Vision | — | 1211 |
| GeoBench | — | 72% |

## Multilingual

- DeepSeek V4 Pro: 54.4 (#45)
- GPT-4.1: 49.4 (#133)

| Benchmark | DeepSeek V4 Pro | GPT-4.1 |
|---|---|---|
| LMArena Non-English | 1439 | 1370 |
| LMArena Chinese | 1486 | 1382 |
| LMArena French | 1472 | 1382 |
| LMArena German | 1458 | 1381 |
| LMArena Japanese | 1445 | 1319 |
| LMArena Korean | 1447 | 1339 |
| LMArena Russian | 1453 | 1377 |
| LMArena Spanish | 1458 | 1376 |

## Instruction Following

- DeepSeek V4 Pro: 76.1 (#47)
- GPT-4.1: 71.3 (#153)

| Benchmark | DeepSeek V4 Pro | GPT-4.1 |
|---|---|---|
| LMArena Instruction Following | 1448 | 1367 |
| IFEval | — | 83.8% |

## Long Context

- DeepSeek V4 Pro: 45.0 (#51)
- GPT-4.1: 40.0 (#163)

| Benchmark | DeepSeek V4 Pro | GPT-4.1 |
|---|---|---|
| LMArena Longer Query | 1458 | 1385 |
| Fiction.LiveBench | — | 63.9% |
| CL-bench Life | 13.5% | — |

## Writing & Preference

- DeepSeek V4 Pro: 65.5 (#46)
- GPT-4.1: 57.6 (#125)

| Benchmark | DeepSeek V4 Pro | GPT-4.1 |
|---|---|---|
| LMArena Text | 1451 | 1383 |
| LMArena Creative Writing | 1446 | 1363 |
| EQ-Bench Creative Writing | 1553 | 1420 |
| LMArena Multi-Turn | 1467 | 1398 |
| WildBench | — | 85.4% |
| EQ-Bench 4 | 1166 | — |

## FAQ

### Is DeepSeek V4 Pro better than GPT-4.1?

DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 35.9 on the Noometry Index.

### Which is cheaper, DeepSeek V4 Pro or GPT-4.1?

DeepSeek V4 Pro is cheaper. It lists at $0.66 per million input tokens and $1.98 per million output tokens; GPT-4.1 lists at $2 and $8.

### Is DeepSeek V4 Pro or GPT-4.1 better for coding?

DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 34.4 in the Noometry coding category.

### Which has the bigger context window?

GPT-4.1 does, with 1.05M tokens against 1M.

### How many benchmarks do DeepSeek V4 Pro and GPT-4.1 share?

34 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and GPT-4.1 has 52.