Model comparison
DeepSeek-V2.5 (Sep 2024) vs Gemini 3.1 Flash Lite
Gemini 3.1 Flash Lite is the stronger model overall, scoring 40.8 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 1 category and Gemini 3.1 Flash Lite in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemini 3.1 Flash Lite leads 60.9 to 49.8.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Gemini 3.1 Flash Lite | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 37.6 | 40.8 |
| Released | 2024-09-06 | 2026-03-03 |
| Weights | Open | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $1.50 |
| Results tracked | 22 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.1 Flash Lite leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Gemini 3.1 Flash Lite: 37.8 (#188)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Coding | 1309 | 1400 |
| Aider Polyglot | 17.8% | — |
| LMArena WebDev | — | 1256 |
| SciCode | — | 41.9% |
| WeirdML | — | 52.2% |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| ALE-Bench | — | 797.73 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Agentic & Tool Use Not comparable
DeepSeek-V2.5 (Sep 2024): —, Gemini 3.1 Flash Lite: 30.2 (#79)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.1 Flash Lite |
|---|---|---|
| DeepResearch Bench | — | 37.3% |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Gemini 3.1 Flash Lite: 22.9 (#186)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1407 |
| Kagi LLM Benchmark | — | 67.2% |
| NYT Connections (extended) | — | 8.2% |
| CritPt | — | 1.1% |
| Chess Puzzles | — | 25% |
| EnigmaEval | — | 3% |
| Thematic Generalization | — | 63.3% |
| DTBench | — | 76.8% |
| LMCA | — | 35% |
| Epoch Capabilities Index | — | 144.47 |
| ForecastBench | — | 54.4 |
Math Gemini 3.1 Flash Lite leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Gemini 3.1 Flash Lite: 40.7 (#90)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Math | 1288 | 1428 |
| FrontierMath (Tiers 1-3) | — | 27.7% |
| OTIS Mock AIME 2024-2025 | — | 80% |
Knowledge Gemini 3.1 Flash Lite leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Gemini 3.1 Flash Lite: 41.9 (#104)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Expert | 1266 | 1398 |
| GPQA Diamond | — | 81.8% |
| Humanity's Last Exam | — | 8.6% |
| Vectara Hallucination Rate | — | 8.2% |
Multimodal Not comparable
DeepSeek-V2.5 (Sep 2024): —, Gemini 3.1 Flash Lite: 39.4 (#60)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Vision | — | 1240 |
Multilingual Gemini 3.1 Flash Lite leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Gemini 3.1 Flash Lite: 52.3 (#86)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Non-English | 1273 | 1411 |
| LMArena Chinese | 1318 | 1461 |
| LMArena French | 1289 | 1424 |
| LMArena German | 1258 | 1429 |
| LMArena Japanese | 1228 | 1413 |
| LMArena Korean | 1209 | 1392 |
| LMArena Russian | 1289 | 1420 |
| LMArena Spanish | 1248 | 1421 |
Instruction Following Gemini 3.1 Flash Lite leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Gemini 3.1 Flash Lite: 72.7 (#131)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Instruction Following | 1280 | 1377 |
Long Context Gemini 3.1 Flash Lite leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Gemini 3.1 Flash Lite: 42.5 (#122)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Longer Query | 1301 | 1394 |
Writing & Preference Gemini 3.1 Flash Lite leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Gemini 3.1 Flash Lite: 60.9 (#94)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Text | 1294 | 1416 |
| LMArena Creative Writing | 1285 | 1401 |
| LMArena Multi-Turn | 1297 | 1417 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Gemini 3.1 Flash Lite?
Gemini 3.1 Flash Lite is the stronger model overall, scoring 40.8 to 37.6 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Gemini 3.1 Flash Lite better for coding?
Gemini 3.1 Flash Lite scores higher on coding benchmarks: 37.8 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Gemini 3.1 Flash Lite share?
17 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Gemini 3.1 Flash Lite has 38.