# DeepSeek-V2.5 (Sep 2024) vs GPT-4o

> DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 28.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v2-5-vs-gpt-4o
- Last updated: 2026-10-10
- Shared benchmarks: 22

## Summary

- They share 22 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 6 categories and GPT-4o in 2 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-V2.5 (Sep 2024) leads 35.9 to 10.6.
- The biggest single-benchmark swing is Aider Polyglot: 17.8% for DeepSeek-V2.5 (Sep 2024) and 45.3% for GPT-4o.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek-V2.5 (Sep 2024) | GPT-4o |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 37.6 | 28.6 |
| Rank | 200 | 324 |
| Context | — | 128K |
| Input $/M | — | $2.50 |
| Output $/M | — | $10 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek-V2.5 (Sep 2024): 31.7 (#281)
- GPT-4o: 24.8 (#328)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-4o |
|---|---|---|
| Aider Polyglot | 17.8% | 45.3% |
| BigCodeBench Instruct | 48.6% | 51.1% |
| LMArena Coding | 1309 | 1297 |
| BigCodeBench Complete | 53.2% | 61.1% |
| HumanEval+ | 83.5% | 87.2% |
| MBPP+ | 74.1% | 72.2% |
| SWE-bench Verified | — | 31% |
| SWE-bench Verified (bash only) | — | 21.6% |
| GSO | — | 0% |
| WeirdML | — | 25.1% |
| LiveBench Coding | — | 51.4% |
| CadEval | — | 26% |

## Agentic & Tool Use

- DeepSeek-V2.5 (Sep 2024): —
- GPT-4o: 21.0 (#141)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-4o |
|---|---|---|
| GDPval | — | 9.9% |
| TheAgentCompany | — | 8.6% |
| Cybench | — | 12.5% |
| BALROG | — | 32.3% |
| LMArena Search | — | 1006 |
| METR Time Horizons | — | 40.8% |

## Reasoning

- DeepSeek-V2.5 (Sep 2024): 25.6 (#145)
- GPT-4o: 9.4 (#343)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-4o |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1281 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 17.8% |
| ARC-AGI-1 | — | 4.5% |
| CritPt | — | 0% |
| Chess Puzzles | — | 13% |
| EnigmaEval | — | 0.8% |
| LiveBench Reasoning | — | 55.8% |
| DTBench | — | 64.5% |
| LiveBench Data Analysis | — | 60.9% |
| LMCA | — | 16.6% |
| Epoch Capabilities Index | — | 128.97 |
| ForecastBench | — | 57.7 |
| LiveBench | — | 55.3% |

## Math

- DeepSeek-V2.5 (Sep 2024): 35.9 (#177)
- GPT-4o: 10.6 (#312)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-4o |
|---|---|---|
| LMArena Math | 1288 | 1285 |
| FrontierMath (Tiers 1-3) | — | 0.4% |
| OTIS Mock AIME 2024-2025 | — | 6.4% |
| Omni-MATH | — | 29.3% |
| LiveBench Math | — | 49.5% |
| MATH Level 5 | — | 53.3% |
| FrontierMath (Feb 2025 set) | — | 0.3% |

## Knowledge

- DeepSeek-V2.5 (Sep 2024): 34.8 (#193)
- GPT-4o: 28.8 (#242)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-4o |
|---|---|---|
| LMArena Expert | 1266 | 1250 |
| GPQA Diamond | — | 49.2% |
| Humanity's Last Exam | — | 2.7% |
| SimpleQA Verified | — | 26% |
| MMLU-Pro | — | 71.3% |
| Confabulations | — | 15.3% |
| Vectara Hallucination Rate | — | 9.6% |
| GPQA (HELM) | — | 52% |
| MMLU | — | 88.1% |

## Multimodal

- DeepSeek-V2.5 (Sep 2024): —
- GPT-4o: 34.5 (#91)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-4o |
|---|---|---|
| LMArena Vision | — | 1137 |
| Video-MME | — | 71.9% |
| GeoBench | — | 71% |
| VPCT | — | 40% |
| ScienceQA | — | 88.5% |

## Multilingual

- DeepSeek-V2.5 (Sep 2024): 42.5 (#193)
- GPT-4o: 43.2 (#186)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-4o |
|---|---|---|
| LMArena Non-English | 1273 | 1283 |
| LMArena Chinese | 1318 | 1277 |
| LMArena French | 1289 | 1304 |
| LMArena German | 1258 | 1282 |
| LMArena Japanese | 1228 | 1257 |
| LMArena Korean | 1209 | 1234 |
| LMArena Russian | 1289 | 1286 |
| LMArena Spanish | 1248 | 1292 |

## Instruction Following

- DeepSeek-V2.5 (Sep 2024): 67.5 (#194)
- GPT-4o: 66.6 (#207)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-4o |
|---|---|---|
| LMArena Instruction Following | 1280 | 1278 |
| LiveBench Instruction Following | — | 68.6% |
| IFEval | — | 81.7% |

## Long Context

- DeepSeek-V2.5 (Sep 2024): 39.5 (#174)
- GPT-4o: 39.4 (#179)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-4o |
|---|---|---|
| LMArena Longer Query | 1301 | 1289 |
| Fiction.LiveBench | — | 66.7% |

## Writing & Preference

- DeepSeek-V2.5 (Sep 2024): 49.8 (#187)
- GPT-4o: 52.6 (#166)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-4o |
|---|---|---|
| LMArena Text | 1294 | 1300 |
| LMArena Creative Writing | 1285 | 1292 |
| LMArena Multi-Turn | 1297 | 1302 |
| Short-Story Creative Writing | — | 81.8% |
| WildBench | — | 82.8% |
| LiveBench Language | — | 47.6% |

## FAQ

### Is DeepSeek-V2.5 (Sep 2024) better than GPT-4o?

DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 28.6 on the Noometry Index.

### Is DeepSeek-V2.5 (Sep 2024) or GPT-4o better for coding?

DeepSeek-V2.5 (Sep 2024) scores higher on coding benchmarks: 31.7 versus 24.8 in the Noometry coding category.

### How many benchmarks do DeepSeek-V2.5 (Sep 2024) and GPT-4o share?

22 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and GPT-4o has 72.