Model comparison
Deepseek Coder v2 vs Gemini 1.5 Pro (May 2024)
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 32.1 on the Noometry Index.
Last verified . 21 shared benchmarks.
DeepSeek
GoogleSummary
- They share 21 benchmarks with published results for both. Deepseek Coder v2 scores higher in 4 categories and Gemini 1.5 Pro (May 2024) in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemini 1.5 Pro (May 2024) leads 52.4 to 38.2.
- Deepseek Coder v2 has downloadable open weights; the other is API-only.
Side by side
| Deepseek Coder v2 | Gemini 1.5 Pro (May 2024) | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 35.9 | 32.1 |
| Released | 2024-06-17 | 2024-02-15 |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 24 | 45 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Deepseek Coder v2 leads
Deepseek Coder v2: 38.1 (#183), Gemini 1.5 Pro (May 2024): 34.2 (#241)
| Benchmark | Deepseek Coder v2 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| BigCodeBench Instruct | 48.2% | 43.8% |
| LMArena Coding | 1251 | 1294 |
| BigCodeBench Complete | 59.7% | 57.5% |
| HumanEval+ | 82.3% | 79.3% |
| MBPP+ | 75.1% | 74.6% |
| WeirdML | — | 22.2% |
| CadEval | — | 34% |
Agentic & Tool Use Not comparable
Deepseek Coder v2: —, Gemini 1.5 Pro (May 2024): 17.9 (#145)
| Benchmark | Deepseek Coder v2 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| TheAgentCompany | — | 3.4% |
| Cybench | — | 7.5% |
| BALROG | — | 21% |
Reasoning Deepseek Coder v2 leads
Deepseek Coder v2: 23.6 (#176), Gemini 1.5 Pro (May 2024): 12.3 (#338)
| Benchmark | Deepseek Coder v2 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1296 |
| ARC-AGI-2 | — | 0.8% |
| SimpleBench | — | 27.1% |
| DTBench | — | 59% |
| BIG-Bench Hard | — | 89.2% |
| Epoch Capabilities Index | — | 131.73 |
| ForecastBench | — | 58.4 |
| WinoGrande | 83.7% | — |
Math Deepseek Coder v2 leads
Deepseek Coder v2: 34.9 (#190), Gemini 1.5 Pro (May 2024): 25.8 (#266)
| Benchmark | Deepseek Coder v2 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Math | 1241 | 1315 |
| OTIS Mock AIME 2024-2025 | — | 23.1% |
| Omni-MATH | — | 36.4% |
| MATH Level 5 | — | 70.4% |
| GSM8K | 94.5% | — |
Knowledge Deepseek Coder v2 leads
Deepseek Coder v2: 32.3 (#212), Gemini 1.5 Pro (May 2024): 29.4 (#239)
| Benchmark | Deepseek Coder v2 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Expert | 1181 | 1279 |
| GPQA Diamond | — | 57.2% |
| Humanity's Last Exam | — | 4.6% |
| MMLU-Pro | — | 73.7% |
| Confabulations | — | 13.5% |
| GPQA (HELM) | — | 53.4% |
| ARC (AI2) Challenge | 64.3% | — |
| MMLU | — | 86.9% |
Multimodal Not comparable
Deepseek Coder v2: —, Gemini 1.5 Pro (May 2024): 36.8 (#77)
| Benchmark | Deepseek Coder v2 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Vision | — | 1161 |
| Video-MME | — | 75% |
Multilingual Gemini 1.5 Pro (May 2024) leads
Deepseek Coder v2: 36.3 (#240), Gemini 1.5 Pro (May 2024): 45.3 (#174)
| Benchmark | Deepseek Coder v2 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Non-English | 1182 | 1312 |
| LMArena Chinese | 1201 | 1331 |
| LMArena French | 1185 | 1302 |
| LMArena German | 1164 | 1286 |
| LMArena Japanese | 1126 | 1292 |
| LMArena Korean | 1104 | 1298 |
| LMArena Russian | 1188 | 1320 |
| LMArena Spanish | 1153 | 1311 |
Instruction Following Gemini 1.5 Pro (May 2024) leads
Deepseek Coder v2: 61.7 (#242), Gemini 1.5 Pro (May 2024): 68.6 (#185)
| Benchmark | Deepseek Coder v2 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Instruction Following | 1180 | 1297 |
| IFEval | — | 83.7% |
Long Context Gemini 1.5 Pro (May 2024) leads
Deepseek Coder v2: 37.0 (#224), Gemini 1.5 Pro (May 2024): 39.8 (#169)
| Benchmark | Deepseek Coder v2 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Longer Query | 1219 | 1308 |
Writing & Preference Gemini 1.5 Pro (May 2024) leads
Deepseek Coder v2: 38.2 (#253), Gemini 1.5 Pro (May 2024): 52.4 (#172)
| Benchmark | Deepseek Coder v2 | Gemini 1.5 Pro (May 2024) |
|---|---|---|
| LMArena Text | 1191 | 1319 |
| LMArena Creative Writing | 1120 | 1333 |
| LMArena Multi-Turn | 1177 | 1296 |
| WildBench | — | 81.3% |
Frequently asked questions
Is Deepseek Coder v2 better than Gemini 1.5 Pro (May 2024)?
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 32.1 on the Noometry Index.
Is Deepseek Coder v2 or Gemini 1.5 Pro (May 2024) better for coding?
Deepseek Coder v2 scores higher on coding benchmarks: 38.1 versus 34.2 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and Gemini 1.5 Pro (May 2024) share?
21 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Gemini 1.5 Pro (May 2024) has 45.