Model comparison
DeepSeek V4 Pro vs Gemini 3.1 Flash Lite
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 40.8 on the Noometry Index. Gemini 3.1 Flash Lite costs 1.8× less per token, which makes it the better buy when DeepSeek V4 Pro's lead doesn't matter for your workload.
Last verified . 33 shared benchmarks.
DeepSeek
GoogleSummary
- They share 33 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 9 categories and Gemini 3.1 Flash Lite in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Pro leads 56.5 to 22.9.
- The biggest single-benchmark swing is NYT Connections (extended): 91.3% for DeepSeek V4 Pro and 8.2% for Gemini 3.1 Flash Lite.
- Gemini 3.1 Flash Lite is cheaper at $0.25 / $1.50 per million input/output tokens, against $0.66 / $1.98 for DeepSeek V4 Pro.
- Gemini 3.1 Flash Lite accepts more context: 1.05M tokens versus 1M.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4 Pro | Gemini 3.1 Flash Lite | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 54.3 | 40.8 |
| Released | 2026-04-24 | 2026-03-03 |
| Weights | Open | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 393K | 66K |
| Input $ / M tokens | $0.66 | $0.25 |
| Output $ / M tokens | $1.98 | $1.50 |
| Results tracked | 48 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Pro leads
DeepSeek V4 Pro: 52.4 (#34), Gemini 3.1 Flash Lite: 37.8 (#188)
| Benchmark | DeepSeek V4 Pro | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena WebDev | 1582 | 1256 |
| SciCode | 51% | 41.9% |
| WeirdML | 66.2% | 52.2% |
| LMArena Coding | 1470 | 1400 |
| ALE-Bench | 1,403 | 797.73 |
| SWE-bench Verified | 77.6% | — |
| FrontierCode | 28.6% | — |
Agentic & Tool Use DeepSeek V4 Pro leads
DeepSeek V4 Pro: 32.8 (#58), Gemini 3.1 Flash Lite: 30.2 (#79)
| Benchmark | DeepSeek V4 Pro | Gemini 3.1 Flash Lite |
|---|---|---|
| APEX-Agents | 47.3% | — |
| DeepResearch Bench | — | 37.3% |
| Vending-Bench 2 | 3,285 | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), Gemini 3.1 Flash Lite: 22.9 (#186)
| Benchmark | DeepSeek V4 Pro | Gemini 3.1 Flash Lite |
|---|---|---|
| Kagi LLM Benchmark | 53.5% | 67.2% |
| NYT Connections (extended) | 91.3% | 8.2% |
| CritPt | 18% | 1.1% |
| Chess Puzzles | 47% | 25% |
| LMArena Hard Prompts | 1461 | 1407 |
| DTBench | 93.9% | 76.8% |
| LMCA | 45.5% | 35% |
| Epoch Capabilities Index | 155.31 | 144.47 |
| ForecastBench | 56.1 | 54.4 |
| ARC-AGI-2 | 61.3% | — |
| ARC-AGI-1 | 90.5% | — |
| EnigmaEval | — | 3% |
| Thematic Generalization | — | 63.3% |
| Mystery Game Puzzles | 43% | — |
| Surface Evolver Bench | 40% | — |
Math DeepSeek V4 Pro leads
DeepSeek V4 Pro: 64.8 (#30), Gemini 3.1 Flash Lite: 40.7 (#90)
| Benchmark | DeepSeek V4 Pro | Gemini 3.1 Flash Lite |
|---|---|---|
| FrontierMath (Tiers 1-3) | 64.6% | 27.7% |
| OTIS Mock AIME 2024-2025 | 98.6% | 80% |
| LMArena Math | 1455 | 1428 |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 76.6% | — |
| ProofBench | 50% | — |
Knowledge DeepSeek V4 Pro leads
DeepSeek V4 Pro: 59.5 (#31), Gemini 3.1 Flash Lite: 41.9 (#104)
| Benchmark | DeepSeek V4 Pro | Gemini 3.1 Flash Lite |
|---|---|---|
| GPQA Diamond | 91.7% | 81.8% |
| Vectara Hallucination Rate | 8.6% | 8.2% |
| LMArena Expert | 1464 | 1398 |
| Humanity's Last Exam | — | 8.6% |
| SimpleQA Verified | 52.9% | — |
Multimodal Not comparable
DeepSeek V4 Pro: —, Gemini 3.1 Flash Lite: 39.4 (#60)
| Benchmark | DeepSeek V4 Pro | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Vision | — | 1240 |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), Gemini 3.1 Flash Lite: 52.3 (#86)
| Benchmark | DeepSeek V4 Pro | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Non-English | 1439 | 1411 |
| LMArena Chinese | 1486 | 1461 |
| LMArena French | 1472 | 1424 |
| LMArena German | 1458 | 1429 |
| LMArena Japanese | 1445 | 1413 |
| LMArena Korean | 1447 | 1392 |
| LMArena Russian | 1453 | 1420 |
| LMArena Spanish | 1458 | 1421 |
Instruction Following DeepSeek V4 Pro leads
DeepSeek V4 Pro: 76.1 (#47), Gemini 3.1 Flash Lite: 72.7 (#131)
| Benchmark | DeepSeek V4 Pro | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Instruction Following | 1448 | 1377 |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), Gemini 3.1 Flash Lite: 42.5 (#122)
| Benchmark | DeepSeek V4 Pro | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Longer Query | 1458 | 1394 |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), Gemini 3.1 Flash Lite: 60.9 (#94)
| Benchmark | DeepSeek V4 Pro | Gemini 3.1 Flash Lite |
|---|---|---|
| LMArena Text | 1451 | 1416 |
| LMArena Creative Writing | 1446 | 1401 |
| LMArena Multi-Turn | 1467 | 1417 |
| EQ-Bench Creative Writing | 1553 | — |
| EQ-Bench 4 | 1166 | — |
Frequently asked questions
Is DeepSeek V4 Pro better than Gemini 3.1 Flash Lite?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 40.8 on the Noometry Index. Gemini 3.1 Flash Lite costs 1.8× less per token, which makes it the better buy when DeepSeek V4 Pro's lead doesn't matter for your workload.
Which is cheaper, DeepSeek V4 Pro or Gemini 3.1 Flash Lite?
Gemini 3.1 Flash Lite is cheaper. It lists at $0.25 per million input tokens and $1.50 per million output tokens; DeepSeek V4 Pro lists at $0.66 and $1.98.
Is DeepSeek V4 Pro or Gemini 3.1 Flash Lite better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 37.8 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.1 Flash Lite does, with 1.05M tokens against 1M.
How many benchmarks do DeepSeek V4 Pro and Gemini 3.1 Flash Lite share?
33 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Gemini 3.1 Flash Lite has 38.