Model comparison
DeepSeek-V3.1 vs Gemini 1.5 Pro (May 2024)
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 32.1 on the Noometry Index.
Last verified . 22 shared benchmarks.
DeepSeek
GoogleSummary
- They share 22 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 7 categories and Gemini 1.5 Pro (May 2024) in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek-V3.1 leads 27.9 to 12.3.
- The biggest single-benchmark swing is DTBench: 82.7% for DeepSeek-V3.1 and 59% for Gemini 1.5 Pro (May 2024).
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.1 | Gemini 1.5 Pro (May 2024) | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 42.8 | 32.1 |
| Released | 2025-08-21 | 2024-02-15 |
| Weights | Open | Proprietary |
| Context window | 164K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $0.95 | — |
| Results tracked | 27 | 45 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
DeepSeek-V3.1: 40.3 (#144), Gemini 1.5 Pro (May 2024): 34.2 (#241)
| Benchmark | DeepSeek-V3.1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| WeirdML | 38.4% | 22.2% |
| LMArena Coding | 1417 | 1294 |
| BigCodeBench Instruct | — | 43.8% |
| BigCodeBench Complete | — | 57.5% |
| CadEval | — | 34% |
| HumanEval+ | — | 79.3% |
| MBPP+ | — | 74.6% |
Agentic & Tool Use Not comparable
DeepSeek-V3.1: —, Gemini 1.5 Pro (May 2024): 17.9 (#145)
| Benchmark | DeepSeek-V3.1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| TheAgentCompany | — | 3.4% |
| Cybench | — | 7.5% |
| BALROG | — | 21% |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.9 (#110), Gemini 1.5 Pro (May 2024): 12.3 (#338)
| Benchmark | DeepSeek-V3.1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| SimpleBench | 40% | 27.1% |
| LMArena Hard Prompts | 1417 | 1296 |
| DTBench | 82.7% | 59% |
| Epoch Capabilities Index | 139.92 | 131.73 |
| ForecastBench | 58 | 58.4 |
| ARC-AGI-2 | — | 0.8% |
| Kagi LLM Benchmark | 53.2% | — |
| LMCA | 24.3% | — |
| BIG-Bench Hard | — | 89.2% |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 38.9 (#122), Gemini 1.5 Pro (May 2024): 25.8 (#266)
| Benchmark | DeepSeek-V3.1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Math | 1420 | 1315 |
| OTIS Mock AIME 2024-2025 | — | 23.1% |
| Omni-MATH | — | 36.4% |
| MATH Level 5 | — | 70.4% |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), Gemini 1.5 Pro (May 2024): 29.4 (#239)
| Benchmark | DeepSeek-V3.1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Expert | 1405 | 1279 |
| GPQA Diamond | — | 57.2% |
| Humanity's Last Exam | — | 4.6% |
| MMLU-Pro | — | 73.7% |
| Confabulations | — | 13.5% |
| Vectara Hallucination Rate | 5.5% | — |
| GPQA (HELM) | — | 53.4% |
| MMLU | — | 86.9% |
Multimodal Not comparable
DeepSeek-V3.1: —, Gemini 1.5 Pro (May 2024): 36.8 (#77)
| Benchmark | DeepSeek-V3.1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Vision | — | 1161 |
| Video-MME | — | 75% |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.6 (#106), Gemini 1.5 Pro (May 2024): 45.3 (#174)
| Benchmark | DeepSeek-V3.1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Non-English | 1400 | 1312 |
| LMArena Chinese | 1469 | 1331 |
| LMArena French | 1447 | 1302 |
| LMArena German | 1411 | 1286 |
| LMArena Japanese | 1378 | 1292 |
| LMArena Korean | 1337 | 1298 |
| LMArena Russian | 1405 | 1320 |
| LMArena Spanish | 1431 | 1311 |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 73.9 (#110), Gemini 1.5 Pro (May 2024): 68.6 (#185)
| Benchmark | DeepSeek-V3.1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Instruction Following | 1400 | 1297 |
| IFEval | — | 83.7% |
Long Context Gemini 1.5 Pro (May 2024) leads
DeepSeek-V3.1: 36.3 (#232), Gemini 1.5 Pro (May 2024): 39.8 (#169)
| Benchmark | DeepSeek-V3.1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Longer Query | 1422 | 1308 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.3 (#98), Gemini 1.5 Pro (May 2024): 52.4 (#172)
| Benchmark | DeepSeek-V3.1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Text | 1420 | 1319 |
| LMArena Creative Writing | 1401 | 1333 |
| LMArena Multi-Turn | 1408 | 1296 |
| EQ-Bench Creative Writing | 1436 | — |
| WildBench | — | 81.3% |
Frequently asked questions
Is DeepSeek-V3.1 better than Gemini 1.5 Pro (May 2024)?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 32.1 on the Noometry Index.
Is DeepSeek-V3.1 or Gemini 1.5 Pro (May 2024) better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 34.2 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1 and Gemini 1.5 Pro (May 2024) share?
22 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Gemini 1.5 Pro (May 2024) has 45.