Model comparison
DeepSeek V4 Pro vs Gemini 1.0 Pro
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 27.3 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 1.0 Pro in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4 Pro leads 64.8 to 9.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.6% for DeepSeek V4 Pro and 1.1% for Gemini 1.0 Pro.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4 Pro | Gemini 1.0 Pro | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 54.3 | 27.3 |
| Released | 2026-04-24 | 2023-12-13 |
| Weights | Open | Proprietary |
| Context window | 1M | — |
| Max output | 393K | — |
| Input $ / M tokens | $0.66 | — |
| Output $ / M tokens | $1.98 | — |
| Results tracked | 48 | 24 |
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 1.0 Pro: 32.2 (#275)
| Benchmark | DeepSeek V4 Pro | Gemini 1.0 Pro |
|---|---|---|
| LMArena Coding | 1470 | 1108 |
| SWE-bench Verified | 77.6% | — |
| FrontierCode | 28.6% | — |
| LMArena WebDev | 1582 | — |
| SciCode | 51% | — |
| WeirdML | 66.2% | — |
| ALE-Bench | 1,403 | — |
| HumanEval+ | — | 55.5% |
| MBPP+ | — | 61.4% |
Agentic & Tool Use Not comparable
DeepSeek V4 Pro: 32.8 (#58), Gemini 1.0 Pro: —
| Benchmark | DeepSeek V4 Pro | Gemini 1.0 Pro |
|---|---|---|
| APEX-Agents | 47.3% | — |
| Vending-Bench 2 | 3,285 | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), Gemini 1.0 Pro: 17.1 (#296)
| Benchmark | DeepSeek V4 Pro | Gemini 1.0 Pro |
|---|---|---|
| LMArena Hard Prompts | 1461 | 1109 |
| DTBench | 93.9% | 45.9% |
| Epoch Capabilities Index | 155.31 | 117.04 |
| 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% | — |
| ForecastBench | 56.1 | — |
Math DeepSeek V4 Pro leads
DeepSeek V4 Pro: 64.8 (#30), Gemini 1.0 Pro: 9.3 (#321)
| Benchmark | DeepSeek V4 Pro | Gemini 1.0 Pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.6% | 1.1% |
| LMArena Math | 1455 | 1132 |
| FrontierMath (Tiers 1-3) | 64.6% | — |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 76.6% | — |
| ProofBench | 50% | — |
| MATH Level 5 | — | 11.2% |
Knowledge DeepSeek V4 Pro leads
DeepSeek V4 Pro: 59.5 (#31), Gemini 1.0 Pro: 15.6 (#291)
| Benchmark | DeepSeek V4 Pro | Gemini 1.0 Pro |
|---|---|---|
| GPQA Diamond | 91.7% | 34% |
| LMArena Expert | 1464 | 1059 |
| SimpleQA Verified | 52.9% | — |
| Vectara Hallucination Rate | 8.6% | — |
| MMLU | — | 70% |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), Gemini 1.0 Pro: 33.4 (#252)
| Benchmark | DeepSeek V4 Pro | Gemini 1.0 Pro |
|---|---|---|
| LMArena Non-English | 1439 | 1138 |
| LMArena Chinese | 1486 | 1124 |
| LMArena French | 1472 | 1145 |
| LMArena German | 1458 | 1125 |
| LMArena Japanese | 1445 | 1023 |
| LMArena Russian | 1453 | 1186 |
| LMArena Spanish | 1458 | 1119 |
| LMArena Korean | 1447 | — |
Instruction Following DeepSeek V4 Pro leads
DeepSeek V4 Pro: 76.1 (#47), Gemini 1.0 Pro: 57.6 (#267)
| Benchmark | DeepSeek V4 Pro | Gemini 1.0 Pro |
|---|---|---|
| LMArena Instruction Following | 1448 | 1114 |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), Gemini 1.0 Pro: 34.3 (#249)
| Benchmark | DeepSeek V4 Pro | Gemini 1.0 Pro |
|---|---|---|
| LMArena Longer Query | 1458 | 1132 |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), Gemini 1.0 Pro: 36.0 (#264)
| Benchmark | DeepSeek V4 Pro | Gemini 1.0 Pro |
|---|---|---|
| LMArena Text | 1451 | 1149 |
| LMArena Creative Writing | 1446 | 1131 |
| LMArena Multi-Turn | 1467 | 1139 |
| EQ-Bench Creative Writing | 1553 | — |
| EQ-Bench 4 | 1166 | — |
Frequently asked questions
Is DeepSeek V4 Pro better than Gemini 1.0 Pro?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 27.3 on the Noometry Index.
Is DeepSeek V4 Pro or Gemini 1.0 Pro better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 32.2 in the Noometry coding category.
How many benchmarks do DeepSeek V4 Pro and Gemini 1.0 Pro share?
20 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Gemini 1.0 Pro has 24.