Model comparison
DeepSeek-V2.5 (Sep 2024) vs Gemini 1.0 Pro
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 27.3 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 7 categories and Gemini 1.0 Pro in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-V2.5 (Sep 2024) leads 35.9 to 9.3.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Gemini 1.0 Pro | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 37.6 | 27.3 |
| Released | 2024-09-06 | 2023-12-13 |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 22 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Gemini 1.0 Pro: 32.2 (#275)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 1.0 Pro |
|---|---|---|
| LMArena Coding | 1309 | 1108 |
| HumanEval+ | 83.5% | 55.5% |
| MBPP+ | 74.1% | 61.4% |
| Aider Polyglot | 17.8% | — |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Gemini 1.0 Pro: 17.1 (#296)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 1.0 Pro |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1109 |
| DTBench | — | 45.9% |
| Epoch Capabilities Index | — | 117.04 |
Math DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Gemini 1.0 Pro: 9.3 (#321)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 1.0 Pro |
|---|---|---|
| LMArena Math | 1288 | 1132 |
| OTIS Mock AIME 2024-2025 | — | 1.1% |
| MATH Level 5 | — | 11.2% |
Knowledge DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Gemini 1.0 Pro: 15.6 (#291)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 1.0 Pro |
|---|---|---|
| LMArena Expert | 1266 | 1059 |
| GPQA Diamond | — | 34% |
| MMLU | — | 70% |
Multilingual DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Gemini 1.0 Pro: 33.4 (#252)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 1.0 Pro |
|---|---|---|
| LMArena Non-English | 1273 | 1138 |
| LMArena Chinese | 1318 | 1124 |
| LMArena French | 1289 | 1145 |
| LMArena German | 1258 | 1125 |
| LMArena Japanese | 1228 | 1023 |
| LMArena Russian | 1289 | 1186 |
| LMArena Spanish | 1248 | 1119 |
| LMArena Korean | 1209 | — |
Instruction Following DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Gemini 1.0 Pro: 57.6 (#267)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 1.0 Pro |
|---|---|---|
| LMArena Instruction Following | 1280 | 1114 |
Long Context DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Gemini 1.0 Pro: 34.3 (#249)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 1.0 Pro |
|---|---|---|
| LMArena Longer Query | 1301 | 1132 |
Writing & Preference DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Gemini 1.0 Pro: 36.0 (#264)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 1.0 Pro |
|---|---|---|
| LMArena Text | 1294 | 1149 |
| LMArena Creative Writing | 1285 | 1131 |
| LMArena Multi-Turn | 1297 | 1139 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Gemini 1.0 Pro?
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 27.3 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Gemini 1.0 Pro better for coding?
They score almost the same on coding (31.7 vs 32.2); test both on your own repository before choosing.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Gemini 1.0 Pro share?
18 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Gemini 1.0 Pro has 24.