# DeepSeek-V3.2-Exp vs Gemini 2.5 Pro

> DeepSeek-V3.2-Exp and Gemini 2.5 Pro score almost the same on the Noometry Index (44.3 vs 45.0), so choose on price, context window or the category you care about most.

- Canonical page: https://noometry.com/compare/deepseek-v3-2-exp-vs-gemini-2-5-pro
- Last updated: 2026-10-10
- Shared benchmarks: 40

## Summary

- They share 40 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 3 categories and Gemini 2.5 Pro in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in long context, where Gemini 2.5 Pro leads 59.8 to 47.6.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 52.2% for DeepSeek-V3.2-Exp and 70.3% for Gemini 2.5 Pro.
- DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 per million input/output tokens, against $1.25 / $10 for Gemini 2.5 Pro.
- Gemini 2.5 Pro accepts more context: 1.05M tokens versus 164K.
- DeepSeek-V3.2-Exp has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek-V3.2-Exp | Gemini 2.5 Pro |
|---|---|---|
| Provider | DeepSeek | Google |
| Noometry Index | 44.3 | 45.0 |
| Rank | 78 | 75 |
| Context | 164K | 1.05M |
| Input $/M | $0.26 | $1.25 |
| Output $/M | $0.38 | $10 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek-V3.2-Exp: 46.5 (#65)
- Gemini 2.5 Pro: 42.4 (#101)

| Benchmark | DeepSeek-V3.2-Exp | Gemini 2.5 Pro |
|---|---|---|
| SWE-bench Verified (bash only) | 70% | 53.6% |
| Aider Polyglot | 74.2% | 83.1% |
| LMArena WebDev | 1362 | 1227 |
| SciCode | 38.9% | 42.8% |
| WeirdML | 39.5% | 54% |
| LMArena Coding | 1454 | 1452 |
| SWE-bench Verified | — | 57.6% |
| SWE-bench Multilingual | 59% | — |
| GSO | — | 3.9% |
| LiveBench Coding | — | 85.9% |
| CadEval | — | 64% |
| ALE-Bench | — | 785.52 |
| AlgoTune | — | 1.51 |

## Agentic & Tool Use

- DeepSeek-V3.2-Exp: 32.7 (#59)
- Gemini 2.5 Pro: 29.2 (#88)

| Benchmark | DeepSeek-V3.2-Exp | Gemini 2.5 Pro |
|---|---|---|
| Terminal-Bench | 39.6% | 32.6% |
| TheAgentCompany | 42.9% | 30.3% |
| Vending-Bench 2 | 1,034 | 573.64 |
| APEX-Agents | 21.3% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| GDPval | — | 23.3% |
| Remote Labor Index | — | 0.8% |
| τ²-bench Banking | — | 13.7% |
| DeepResearch Bench | — | 42.8% |
| BALROG | — | 43.3% |
| LMArena Search | — | 1142 |
| METR Time Horizons | — | 55.4% |

## Reasoning

- DeepSeek-V3.2-Exp: 22.1 (#208)
- Gemini 2.5 Pro: 28.8 (#99)

| Benchmark | DeepSeek-V3.2-Exp | Gemini 2.5 Pro |
|---|---|---|
| ARC-AGI-2 | 4% | 4.9% |
| Kagi LLM Benchmark | 52.2% | 70.3% |
| ARC-AGI-1 | 57% | 41% |
| CritPt | 2.9% | 2% |
| Chess Puzzles | 14% | 20% |
| LMArena Hard Prompts | 1434 | 1455 |
| DTBench | 87.7% | 82.4% |
| LMCA | 29.1% | 34.8% |
| Epoch Capabilities Index | 146.27 | 145.32 |
| SimpleBench | — | 62.4% |
| NYT Connections (extended) | 36.7% | — |
| EnigmaEval | — | 5.6% |
| Thematic Generalization | 65% | — |
| LiveBench Reasoning | — | 89.8% |
| LiveBench Data Analysis | — | 79.9% |
| ForecastBench | — | 61.3 |
| LiveBench | — | 82.3% |

## Math

- DeepSeek-V3.2-Exp: 41.7 (#87)
- Gemini 2.5 Pro: 32.5 (#213)

| Benchmark | DeepSeek-V3.2-Exp | Gemini 2.5 Pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 84.7% |
| LMArena Math | 1435 | 1450 |
| FrontierMath (Feb 2025 set) | 22.1% | 14.1% |
| FrontierMath Tier 4 (v1) | 2.1% | 4.2% |
| FrontierMath (Tiers 1-3) | — | 24.6% |
| FrontierMath Tier 4 | — | 0% |
| MathArena Final-Answer Competitions | 57.7% | — |
| ProofBench | 8% | — |
| Omni-MATH | — | 41.6% |
| LiveBench Math | — | 90.2% |
| MATH Level 5 | — | 95.9% |

## Knowledge

- DeepSeek-V3.2-Exp: 51.7 (#66)
- Gemini 2.5 Pro: 56.0 (#46)

| Benchmark | DeepSeek-V3.2-Exp | Gemini 2.5 Pro |
|---|---|---|
| GPQA Diamond | 83.4% | 85.3% |
| Vectara Hallucination Rate | 5.3% | 7% |
| LMArena Expert | 1436 | 1452 |
| Humanity's Last Exam | — | 21.6% |
| MMLU-Pro | — | 86.3% |
| Confabulations | — | 10.6% |
| GPQA (HELM) | — | 74.9% |

## Multimodal

- DeepSeek-V3.2-Exp: —
- Gemini 2.5 Pro: 45.2 (#18)

| Benchmark | DeepSeek-V3.2-Exp | Gemini 2.5 Pro |
|---|---|---|
| LMArena Vision | — | 1263 |
| GeoBench | — | 86% |
| VPCT | — | 48% |
| LMArena Document | — | 1421 |
| SpatialViz-Bench | — | 44.7% |

## Multilingual

- DeepSeek-V3.2-Exp: 52.2 (#90)
- Gemini 2.5 Pro: 55.3 (#31)

| Benchmark | DeepSeek-V3.2-Exp | Gemini 2.5 Pro |
|---|---|---|
| LMArena Non-English | 1409 | 1451 |
| LMArena Chinese | 1461 | 1507 |
| LMArena French | 1433 | 1472 |
| LMArena German | 1440 | 1487 |
| LMArena Japanese | 1374 | 1461 |
| LMArena Korean | 1371 | 1434 |
| LMArena Russian | 1424 | 1461 |
| LMArena Spanish | 1440 | 1473 |

## Instruction Following

- DeepSeek-V3.2-Exp: 74.5 (#93)
- Gemini 2.5 Pro: 75.0 (#75)

| Benchmark | DeepSeek-V3.2-Exp | Gemini 2.5 Pro |
|---|---|---|
| LMArena Instruction Following | 1413 | 1437 |
| LiveBench Instruction Following | — | 80.6% |
| IFEval | — | 84% |

## Long Context

- DeepSeek-V3.2-Exp: 47.6 (#16)
- Gemini 2.5 Pro: 59.8 (#5)

| Benchmark | DeepSeek-V3.2-Exp | Gemini 2.5 Pro |
|---|---|---|
| Fiction.LiveBench | 83.3% | 91.7% |
| LMArena Longer Query | 1428 | 1449 |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |

## Writing & Preference

- DeepSeek-V3.2-Exp: 62.4 (#77)
- Gemini 2.5 Pro: 63.7 (#62)

| Benchmark | DeepSeek-V3.2-Exp | Gemini 2.5 Pro |
|---|---|---|
| LMArena Text | 1425 | 1458 |
| LMArena Creative Writing | 1403 | 1454 |
| EQ-Bench Creative Writing | 1515 | 1421 |
| LMArena Multi-Turn | 1427 | 1453 |
| Short-Story Creative Writing | — | 83.8% |
| WildBench | — | 85.7% |
| LiveBench Language | — | 67.8% |

## FAQ

### Is DeepSeek-V3.2-Exp better than Gemini 2.5 Pro?

DeepSeek-V3.2-Exp and Gemini 2.5 Pro score almost the same on the Noometry Index (44.3 vs 45.0), so choose on price, context window or the category you care about most.

### Which is cheaper, DeepSeek-V3.2-Exp or Gemini 2.5 Pro?

DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; Gemini 2.5 Pro lists at $1.25 and $10.

### Is DeepSeek-V3.2-Exp or Gemini 2.5 Pro better for coding?

DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.5 versus 42.4 in the Noometry coding category.

### Which has the bigger context window?

Gemini 2.5 Pro does, with 1.05M tokens against 164K.

### How many benchmarks do DeepSeek-V3.2-Exp and Gemini 2.5 Pro share?

40 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 49 scored results on Noometry and Gemini 2.5 Pro has 78.