Model comparison
DeepSeek-V2.5 (Sep 2024) vs Gemini 2.5 Pro
Gemini 2.5 Pro is the stronger model overall, scoring 45.0 to 37.6 on the Noometry Index.
Last verified . 18 shared benchmarks.
DeepSeek
GoogleSummary
- They share 18 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 1 category and Gemini 2.5 Pro in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 2.5 Pro leads 56.0 to 34.8.
- The biggest single-benchmark swing is Aider Polyglot: 17.8% for DeepSeek-V2.5 (Sep 2024) and 83.1% for Gemini 2.5 Pro.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Gemini 2.5 Pro | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 37.6 | 45.0 |
| Released | 2024-09-06 | 2025-03-25 |
| Weights | Open | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 66K |
| Input $ / M tokens | — | $1.25 |
| Output $ / M tokens | — | $10 |
| Results tracked | 22 | 78 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 2.5 Pro leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Gemini 2.5 Pro: 42.4 (#101)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 2.5 Pro |
|---|---|---|
| Aider Polyglot | 17.8% | 83.1% |
| LMArena Coding | 1309 | 1452 |
| SWE-bench Verified | — | 57.6% |
| SWE-bench Verified (bash only) | — | 53.6% |
| LMArena WebDev | — | 1227 |
| SciCode | — | 42.8% |
| GSO | — | 3.9% |
| WeirdML | — | 54% |
| BigCodeBench Instruct | 48.6% | — |
| LiveBench Coding | — | 85.9% |
| BigCodeBench Complete | 53.2% | — |
| CadEval | — | 64% |
| ALE-Bench | — | 785.52 |
| AlgoTune | — | 1.51 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Agentic & Tool Use Not comparable
DeepSeek-V2.5 (Sep 2024): —, Gemini 2.5 Pro: 29.2 (#88)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 2.5 Pro |
|---|---|---|
| Terminal-Bench | — | 32.6% |
| GDPval | — | 23.3% |
| Remote Labor Index | — | 0.8% |
| TheAgentCompany | — | 30.3% |
| τ²-bench Banking | — | 13.7% |
| DeepResearch Bench | — | 42.8% |
| BALROG | — | 43.3% |
| LMArena Search | — | 1142 |
| METR Time Horizons | — | 55.4% |
| Vending-Bench 2 | — | 573.64 |
Reasoning Gemini 2.5 Pro leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Gemini 2.5 Pro: 28.8 (#99)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 2.5 Pro |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1455 |
| ARC-AGI-2 | — | 4.9% |
| SimpleBench | — | 62.4% |
| Kagi LLM Benchmark | — | 70.3% |
| ARC-AGI-1 | — | 41% |
| CritPt | — | 2% |
| Chess Puzzles | — | 20% |
| EnigmaEval | — | 5.6% |
| LiveBench Reasoning | — | 89.8% |
| DTBench | — | 82.4% |
| LiveBench Data Analysis | — | 79.9% |
| LMCA | — | 34.8% |
| Epoch Capabilities Index | — | 145.32 |
| ForecastBench | — | 61.3 |
| LiveBench | — | 82.3% |
Math DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Gemini 2.5 Pro: 32.5 (#213)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 2.5 Pro |
|---|---|---|
| LMArena Math | 1288 | 1450 |
| FrontierMath (Tiers 1-3) | — | 24.6% |
| FrontierMath Tier 4 | — | 0% |
| OTIS Mock AIME 2024-2025 | — | 84.7% |
| Omni-MATH | — | 41.6% |
| LiveBench Math | — | 90.2% |
| MATH Level 5 | — | 95.9% |
| FrontierMath (Feb 2025 set) | — | 14.1% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Gemini 2.5 Pro leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Gemini 2.5 Pro: 56.0 (#46)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 2.5 Pro |
|---|---|---|
| LMArena Expert | 1266 | 1452 |
| GPQA Diamond | — | 85.3% |
| Humanity's Last Exam | — | 21.6% |
| MMLU-Pro | — | 86.3% |
| Confabulations | — | 10.6% |
| Vectara Hallucination Rate | — | 7% |
| GPQA (HELM) | — | 74.9% |
Multimodal Not comparable
DeepSeek-V2.5 (Sep 2024): —, Gemini 2.5 Pro: 45.2 (#18)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 2.5 Pro |
|---|---|---|
| LMArena Vision | — | 1263 |
| GeoBench | — | 86% |
| VPCT | — | 48% |
| LMArena Document | — | 1421 |
| SpatialViz-Bench | — | 44.7% |
Multilingual Gemini 2.5 Pro leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Gemini 2.5 Pro: 55.3 (#31)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 2.5 Pro |
|---|---|---|
| LMArena Non-English | 1273 | 1451 |
| LMArena Chinese | 1318 | 1507 |
| LMArena French | 1289 | 1472 |
| LMArena German | 1258 | 1487 |
| LMArena Japanese | 1228 | 1461 |
| LMArena Korean | 1209 | 1434 |
| LMArena Russian | 1289 | 1461 |
| LMArena Spanish | 1248 | 1473 |
Instruction Following Gemini 2.5 Pro leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Gemini 2.5 Pro: 75.0 (#75)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 2.5 Pro |
|---|---|---|
| LMArena Instruction Following | 1280 | 1437 |
| LiveBench Instruction Following | — | 80.6% |
| IFEval | — | 84% |
Long Context Gemini 2.5 Pro leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Gemini 2.5 Pro: 59.8 (#5)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 2.5 Pro |
|---|---|---|
| LMArena Longer Query | 1301 | 1449 |
| Fiction.LiveBench | — | 91.7% |
Writing & Preference Gemini 2.5 Pro leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Gemini 2.5 Pro: 63.7 (#62)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 2.5 Pro |
|---|---|---|
| LMArena Text | 1294 | 1458 |
| LMArena Creative Writing | 1285 | 1454 |
| LMArena Multi-Turn | 1297 | 1453 |
| Short-Story Creative Writing | — | 83.8% |
| EQ-Bench Creative Writing | — | 1421 |
| WildBench | — | 85.7% |
| LiveBench Language | — | 67.8% |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Gemini 2.5 Pro?
Gemini 2.5 Pro is the stronger model overall, scoring 45.0 to 37.6 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Gemini 2.5 Pro better for coding?
Gemini 2.5 Pro scores higher on coding benchmarks: 42.4 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Gemini 2.5 Pro share?
18 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Gemini 2.5 Pro has 78.