Model comparison
DeepSeek-R1 vs Gemini 1.5 Pro (May 2024)
DeepSeek-R1 is the stronger model overall, scoring 42.3 to 32.1 on the Noometry Index.
Last verified . 32 shared benchmarks.
DeepSeek
GoogleSummary
- They share 32 benchmarks with published results for both. DeepSeek-R1 scores higher in 9 categories and Gemini 1.5 Pro (May 2024) in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-R1 leads 43.8 to 25.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 66.4% for DeepSeek-R1 and 23.1% for Gemini 1.5 Pro (May 2024).
Side by side
| DeepSeek-R1 | Gemini 1.5 Pro (May 2024) | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 42.3 | 32.1 |
| Released | 2025-01-20 | 2024-02-15 |
| Weights | Proprietary | Proprietary |
| Context window | 164K | — |
| Max output | 64K | — |
| Input $ / M tokens | $0.50 | — |
| Output $ / M tokens | $2.15 | — |
| Results tracked | 52 | 45 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1 leads
DeepSeek-R1: 46.3 (#68), Gemini 1.5 Pro (May 2024): 34.2 (#241)
| Benchmark | DeepSeek-R1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| WeirdML | 41.6% | 22.2% |
| LMArena Coding | 1427 | 1294 |
| Aider Polyglot | 71.4% | — |
| SciCode | 35.7% | — |
| BigCodeBench Instruct | — | 43.8% |
| LiveBench Coding | 66.7% | — |
| BigCodeBench Complete | — | 57.5% |
| CadEval | — | 34% |
| ALE-Bench | 804.12 | — |
| AlgoTune | 1.7 | — |
| HumanEval+ | — | 79.3% |
| MBPP+ | — | 74.6% |
Agentic & Tool Use DeepSeek-R1 leads
DeepSeek-R1: 30.7 (#75), Gemini 1.5 Pro (May 2024): 17.9 (#145)
| Benchmark | DeepSeek-R1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| BALROG | 34.9% | 21% |
| TheAgentCompany | — | 3.4% |
| Cybench | — | 7.5% |
| DeepResearch Bench | 35.1% | — |
| METR Time Horizons | 53.8% | — |
Reasoning DeepSeek-R1 leads
DeepSeek-R1: 18.6 (#278), Gemini 1.5 Pro (May 2024): 12.3 (#338)
| Benchmark | DeepSeek-R1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| ARC-AGI-2 | 1.3% | 0.8% |
| SimpleBench | 40.8% | 27.1% |
| LMArena Hard Prompts | 1416 | 1296 |
| Epoch Capabilities Index | 141.29 | 131.73 |
| ForecastBench | 60 | 58.4 |
| Kagi LLM Benchmark | 69.4% | — |
| ARC-AGI-1 | 21.2% | — |
| CritPt | 1.1% | — |
| LiveBench Reasoning | 83.2% | — |
| DTBench | — | 59% |
| LiveBench Data Analysis | 69.8% | — |
| BIG-Bench Hard | — | 89.2% |
| LiveBench | 71.6% | — |
Math DeepSeek-R1 leads
DeepSeek-R1: 43.8 (#79), Gemini 1.5 Pro (May 2024): 25.8 (#266)
| Benchmark | DeepSeek-R1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.4% | 23.1% |
| Omni-MATH | 42.4% | 36.4% |
| LMArena Math | 1400 | 1315 |
| MATH Level 5 | 96.6% | 70.4% |
| LiveBench Math | 80.7% | — |
Knowledge DeepSeek-R1 leads
DeepSeek-R1: 44.5 (#87), Gemini 1.5 Pro (May 2024): 29.4 (#239)
| Benchmark | DeepSeek-R1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| GPQA Diamond | 76.3% | 57.2% |
| MMLU-Pro | 79.3% | 73.7% |
| Confabulations | 12.7% | 13.5% |
| GPQA (HELM) | 66.6% | 53.4% |
| LMArena Expert | 1394 | 1279 |
| Humanity's Last Exam | — | 4.6% |
| Vectara Hallucination Rate | 11.3% | — |
| MMLU | — | 86.9% |
Multimodal Not comparable
DeepSeek-R1: —, Gemini 1.5 Pro (May 2024): 36.8 (#77)
| Benchmark | DeepSeek-R1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Vision | — | 1161 |
| Video-MME | — | 75% |
Multilingual DeepSeek-R1 leads
DeepSeek-R1: 52.4 (#85), Gemini 1.5 Pro (May 2024): 45.3 (#174)
| Benchmark | DeepSeek-R1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Non-English | 1412 | 1312 |
| LMArena Chinese | 1442 | 1331 |
| LMArena French | 1417 | 1302 |
| LMArena German | 1404 | 1286 |
| LMArena Japanese | 1391 | 1292 |
| LMArena Korean | 1360 | 1298 |
| LMArena Russian | 1423 | 1320 |
| LMArena Spanish | 1411 | 1311 |
Instruction Following DeepSeek-R1 leads
DeepSeek-R1: 72.0 (#143), Gemini 1.5 Pro (May 2024): 68.6 (#185)
| Benchmark | DeepSeek-R1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| IFEval | 78.4% | 83.7% |
| LMArena Instruction Following | 1382 | 1297 |
| LiveBench Instruction Following | 80.5% | — |
Long Context DeepSeek-R1 leads
DeepSeek-R1: 45.4 (#36), Gemini 1.5 Pro (May 2024): 39.8 (#169)
| Benchmark | DeepSeek-R1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Longer Query | 1391 | 1308 |
| Fiction.LiveBench | 75% | — |
Writing & Preference DeepSeek-R1 leads
DeepSeek-R1: 61.4 (#88), Gemini 1.5 Pro (May 2024): 52.4 (#172)
| Benchmark | DeepSeek-R1 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Text | 1428 | 1319 |
| LMArena Creative Writing | 1405 | 1333 |
| WildBench | 82.8% | 81.3% |
| LMArena Multi-Turn | 1405 | 1296 |
| Short-Story Creative Writing | 83% | — |
| EQ-Bench Creative Writing | 1500 | — |
| LiveBench Language | 48.5% | — |
Frequently asked questions
Is DeepSeek-R1 better than Gemini 1.5 Pro (May 2024)?
DeepSeek-R1 is the stronger model overall, scoring 42.3 to 32.1 on the Noometry Index.
Is DeepSeek-R1 or Gemini 1.5 Pro (May 2024) better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 46.3 versus 34.2 in the Noometry coding category.
How many benchmarks do DeepSeek-R1 and Gemini 1.5 Pro (May 2024) share?
32 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and Gemini 1.5 Pro (May 2024) has 45.