Model comparison
DeepSeek-V2.5 (Sep 2024) vs Gemini 3.7 Flash
Gemini 3.7 Flash is the stronger model overall, scoring 59.8 to 37.6 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeek
GoogleSummary
- They share 17 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 0 categories and Gemini 3.7 Flash in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.7 Flash leads 70.0 to 25.6.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Gemini 3.7 Flash | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 37.6 | 59.8 |
| Released | 2024-09-06 | 2026-08-13 |
| Weights | Open | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.75 |
| Output $ / M tokens | — | $3.75 |
| Results tracked | 22 | 44 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.7 Flash leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Gemini 3.7 Flash: 56.2 (#22)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.7 Flash |
|---|---|---|
| LMArena Coding | 1309 | 1497 |
| DeepSWE | — | 65.5% |
| FrontierCode | — | 43.6% |
| Aider Polyglot | 17.8% | — |
| LMArena WebDev | — | 1592 |
| FrontierSWE | — | 20.3% |
| SciCode | — | 59.8% |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| ALE-Bench | — | 904.3 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Agentic & Tool Use Not comparable
DeepSeek-V2.5 (Sep 2024): —, Gemini 3.7 Flash: 42.1 (#19)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.7 Flash |
|---|---|---|
| APEX-Agents | — | 67.8% |
| Remote Labor Index | — | 5% |
| GDP.pdf | — | 23.8% |
Reasoning Gemini 3.7 Flash leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Gemini 3.7 Flash: 70.0 (#15)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.7 Flash |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1494 |
| ARC-AGI-2 | — | 84.6% |
| NYT Connections (extended) | — | 94% |
| ARC-AGI-1 | — | 95.5% |
| CritPt | — | 14.3% |
| Chess Puzzles | — | 47% |
| Mystery Game Puzzles | — | 37% |
| DTBench | — | 96.8% |
| LMCA | — | 50.4% |
| Epoch Capabilities Index | — | 157.27 |
Math Gemini 3.7 Flash leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Gemini 3.7 Flash: 69.6 (#23)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.7 Flash |
|---|---|---|
| LMArena Math | 1288 | 1507 |
| FrontierMath (Tiers 1-3) | — | 71.6% |
| FrontierMath Tier 4 | — | 36.6% |
| OTIS Mock AIME 2024-2025 | — | 97.2% |
| ProofBench | — | 58% |
Knowledge Gemini 3.7 Flash leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Gemini 3.7 Flash: 69.7 (#5)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.7 Flash |
|---|---|---|
| LMArena Expert | 1266 | 1508 |
| GPQA Diamond | — | 94.8% |
| SimpleQA Verified | — | 69.2% |
Multimodal Not comparable
DeepSeek-V2.5 (Sep 2024): —, Gemini 3.7 Flash: 37.3 (#73)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.7 Flash |
|---|---|---|
| LMArena Vision | — | 1316 |
| Furniture Assembly | — | 26.7% |
Multilingual Gemini 3.7 Flash leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Gemini 3.7 Flash: 57.6 (#7)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.7 Flash |
|---|---|---|
| LMArena Non-English | 1273 | 1484 |
| LMArena Chinese | 1318 | 1548 |
| LMArena French | 1289 | 1505 |
| LMArena German | 1258 | 1498 |
| LMArena Japanese | 1228 | 1512 |
| LMArena Korean | 1209 | 1483 |
| LMArena Russian | 1289 | 1516 |
| LMArena Spanish | 1248 | 1503 |
Instruction Following Gemini 3.7 Flash leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Gemini 3.7 Flash: 77.7 (#15)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.7 Flash |
|---|---|---|
| LMArena Instruction Following | 1280 | 1483 |
Long Context Gemini 3.7 Flash leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Gemini 3.7 Flash: 45.7 (#30)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.7 Flash |
|---|---|---|
| LMArena Longer Query | 1301 | 1492 |
Writing & Preference Gemini 3.7 Flash leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Gemini 3.7 Flash: 71.2 (#20)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.7 Flash |
|---|---|---|
| LMArena Text | 1294 | 1486 |
| LMArena Creative Writing | 1285 | 1490 |
| LMArena Multi-Turn | 1297 | 1489 |
| EQ-Bench Creative Writing | — | 1723 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Gemini 3.7 Flash?
Gemini 3.7 Flash is the stronger model overall, scoring 59.8 to 37.6 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Gemini 3.7 Flash better for coding?
Gemini 3.7 Flash scores higher on coding benchmarks: 56.2 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Gemini 3.7 Flash share?
17 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Gemini 3.7 Flash has 44.