Model comparison
DeepSeek V4 Pro vs Gemini 1.5 Flash 8B
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 29.9 on the Noometry Index.
Last verified . 20 shared benchmarks.
DeepSeek
GoogleSummary
- They share 20 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 8 categories and Gemini 1.5 Flash 8B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4 Pro leads 64.8 to 14.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.6% for DeepSeek V4 Pro and 4.6% for Gemini 1.5 Flash 8B.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4 Pro | Gemini 1.5 Flash 8B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 54.3 | 29.9 |
| Released | 2026-04-24 | 2024-10-03 |
| Weights | Open | Proprietary |
| Context window | 1M | — |
| Max output | 393K | — |
| Input $ / M tokens | $0.66 | — |
| Output $ / M tokens | $1.98 | — |
| Results tracked | 48 | 21 |
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 1.5 Flash 8B: 35.5 (#225)
| Benchmark | DeepSeek V4 Pro | Gemini 1.5 Flash 8B |
|---|---|---|
| LMArena Coding | 1470 | 1218 |
| SWE-bench Verified | 77.6% | — |
| FrontierCode | 28.6% | — |
| LMArena WebDev | 1582 | — |
| SciCode | 51% | — |
| WeirdML | 66.2% | — |
| ALE-Bench | 1,403 | — |
Agentic & Tool Use Not comparable
DeepSeek V4 Pro: 32.8 (#58), Gemini 1.5 Flash 8B: —
| Benchmark | DeepSeek V4 Pro | Gemini 1.5 Flash 8B |
|---|---|---|
| APEX-Agents | 47.3% | — |
| Vending-Bench 2 | 3,285 | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), Gemini 1.5 Flash 8B: 20.0 (#244)
| Benchmark | DeepSeek V4 Pro | Gemini 1.5 Flash 8B |
|---|---|---|
| LMArena Hard Prompts | 1461 | 1209 |
| DTBench | 93.9% | 50% |
| ARC-AGI-2 | 61.3% | — |
| Kagi LLM Benchmark | 53.5% | — |
| NYT Connections (extended) | 91.3% | — |
| ARC-AGI-1 | 90.5% | — |
| CritPt | 18% | — |
| Chess Puzzles | 47% | — |
| Mystery Game Puzzles | 43% | — |
| LMCA | 45.5% | — |
| Surface Evolver Bench | 40% | — |
| Epoch Capabilities Index | 155.31 | — |
| ForecastBench | 56.1 | — |
Math DeepSeek V4 Pro leads
DeepSeek V4 Pro: 64.8 (#30), Gemini 1.5 Flash 8B: 14.2 (#302)
| Benchmark | DeepSeek V4 Pro | Gemini 1.5 Flash 8B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.6% | 4.6% |
| LMArena Math | 1455 | 1207 |
| FrontierMath (Tiers 1-3) | 64.6% | — |
| 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 1.5 Flash 8B: 16.0 (#289)
| Benchmark | DeepSeek V4 Pro | Gemini 1.5 Flash 8B |
|---|---|---|
| GPQA Diamond | 91.7% | 33% |
| LMArena Expert | 1464 | 1185 |
| SimpleQA Verified | 52.9% | — |
| Vectara Hallucination Rate | 8.6% | — |
Multimodal Not comparable
DeepSeek V4 Pro: —, Gemini 1.5 Flash 8B: 28.2 (#115)
| Benchmark | DeepSeek V4 Pro | Gemini 1.5 Flash 8B |
|---|---|---|
| LMArena Vision | — | 1044 |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), Gemini 1.5 Flash 8B: 38.5 (#229)
| Benchmark | DeepSeek V4 Pro | Gemini 1.5 Flash 8B |
|---|---|---|
| LMArena Non-English | 1439 | 1215 |
| LMArena Chinese | 1486 | 1231 |
| LMArena French | 1472 | 1234 |
| LMArena German | 1458 | 1206 |
| LMArena Japanese | 1445 | 1150 |
| LMArena Korean | 1447 | 1140 |
| LMArena Russian | 1453 | 1236 |
| LMArena Spanish | 1458 | 1212 |
Instruction Following DeepSeek V4 Pro leads
DeepSeek V4 Pro: 76.1 (#47), Gemini 1.5 Flash 8B: 62.8 (#236)
| Benchmark | DeepSeek V4 Pro | Gemini 1.5 Flash 8B |
|---|---|---|
| LMArena Instruction Following | 1448 | 1199 |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), Gemini 1.5 Flash 8B: 37.0 (#225)
| Benchmark | DeepSeek V4 Pro | Gemini 1.5 Flash 8B |
|---|---|---|
| LMArena Longer Query | 1458 | 1219 |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), Gemini 1.5 Flash 8B: 42.8 (#232)
| Benchmark | DeepSeek V4 Pro | Gemini 1.5 Flash 8B |
|---|---|---|
| LMArena Text | 1451 | 1226 |
| LMArena Creative Writing | 1446 | 1218 |
| LMArena Multi-Turn | 1467 | 1185 |
| EQ-Bench Creative Writing | 1553 | — |
| EQ-Bench 4 | 1166 | — |
Frequently asked questions
Is DeepSeek V4 Pro better than Gemini 1.5 Flash 8B?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 29.9 on the Noometry Index.
Is DeepSeek V4 Pro or Gemini 1.5 Flash 8B better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 35.5 in the Noometry coding category.
How many benchmarks do DeepSeek V4 Pro and Gemini 1.5 Flash 8B share?
20 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Gemini 1.5 Flash 8B has 21.