Model comparison
DeepSeek-V3.1-Terminus vs Gemini 3 Pro
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 43.1 on the Noometry Index.
Last verified . 13 shared benchmarks.
DeepSeek
GoogleSummary
- They share 13 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 0 categories and Gemini 3 Pro in 7 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3 Pro leads 52.5 to 26.4.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 57.4% for DeepSeek-V3.1-Terminus and 80.1% for Gemini 3 Pro.
- DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.1-Terminus | Gemini 3 Pro | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 43.1 | 54.8 |
| Released | 2025-09-22 | 2025-11-18 |
| Weights | Open | Proprietary |
| Context window | 164K | — |
| Max output | 147K | — |
| Input $ / M tokens | $0.27 | — |
| Output $ / M tokens | $1 | — |
| Results tracked | 16 | 67 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3 Pro leads
DeepSeek-V3.1-Terminus: 42.0 (#113), Gemini 3 Pro: 51.6 (#39)
| Benchmark | DeepSeek-V3.1-Terminus | Gemini 3 Pro |
|---|---|---|
| LMArena Coding | 1426 | 1481 |
| ALE-Bench | 745.17 | 1,177 |
| SWE-bench Verified | — | 72.9% |
| SWE-bench Verified (bash only) | — | 74.2% |
| LMArena WebDev | — | 1440 |
| SWE-bench Multilingual | — | 68.7% |
| SciCode | 40.6% | — |
| GSO | — | 18.6% |
| WeirdML | — | 69.9% |
| AlgoTune | — | 1.83 |
Agentic & Tool Use Not comparable
DeepSeek-V3.1-Terminus: —, Gemini 3 Pro: 40.6 (#23)
| Benchmark | DeepSeek-V3.1-Terminus | Gemini 3 Pro |
|---|---|---|
| Terminal-Bench | — | 69.4% |
| Berkeley Function Calling Leaderboard | — | 72.5% |
| GDPval | — | 40.3% |
| Remote Labor Index | — | 1.3% |
| τ²-bench Airline | — | 80.5% |
| τ²-bench Banking | — | 18% |
| τ²-bench Retail | — | 75.9% |
| τ²-bench Telecom | — | 91% |
| DeepResearch Bench | — | 46.3% |
| BALROG | — | 58.1% |
| LMArena Search | — | 1207 |
| METR Time Horizons | — | 71% |
| Vending-Bench 2 | — | 5,478 |
Reasoning Gemini 3 Pro leads
DeepSeek-V3.1-Terminus: 26.4 (#133), Gemini 3 Pro: 52.5 (#31)
| Benchmark | DeepSeek-V3.1-Terminus | Gemini 3 Pro |
|---|---|---|
| Kagi LLM Benchmark | 57.4% | 80.1% |
| CritPt | 1.7% | 6.9% |
| LMArena Hard Prompts | 1426 | 1480 |
| ARC-AGI-2 | — | 31.1% |
| SimpleBench | — | 76.4% |
| NYT Connections (extended) | — | 94.4% |
| ARC-AGI-1 | — | 75% |
| Chess Puzzles | — | 31% |
| EnigmaEval | — | 18.2% |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |
| Epoch Capabilities Index | — | 152.92 |
| ForecastBench | — | 61.2 |
Math Gemini 3 Pro leads
DeepSeek-V3.1-Terminus: 38.5 (#137), Gemini 3 Pro: 49.9 (#59)
| Benchmark | DeepSeek-V3.1-Terminus | Gemini 3 Pro |
|---|---|---|
| LMArena Math | 1402 | 1476 |
| MathArena Final-Answer Competitions | — | 67% |
| OTIS Mock AIME 2024-2025 | — | 91.4% |
| ProofBench | — | 20% |
| Omni-MATH | — | 55.5% |
| FrontierMath (Feb 2025 set) | — | 37.6% |
| FrontierMath Tier 4 (v1) | — | 18.8% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Gemini 3 Pro: 64.4 (#16)
| Benchmark | DeepSeek-V3.1-Terminus | Gemini 3 Pro |
|---|---|---|
| GPQA Diamond | — | 92.6% |
| Humanity's Last Exam | — | 37.5% |
| MMLU-Pro | — | 90.3% |
| Vectara Hallucination Rate | — | 13.6% |
| GPQA (HELM) | — | 80.3% |
| LMArena Expert | — | 1475 |
Multimodal Not comparable
DeepSeek-V3.1-Terminus: —, Gemini 3 Pro: 57.6 (#2)
| Benchmark | DeepSeek-V3.1-Terminus | Gemini 3 Pro |
|---|---|---|
| LMArena Vision | — | 1305 |
| GeoBench | — | 84% |
| VPCT | — | 91% |
| LMArena Document | — | 1434 |
Multilingual Gemini 3 Pro leads
DeepSeek-V3.1-Terminus: 52.1 (#92), Gemini 3 Pro: 56.9 (#16)
| Benchmark | DeepSeek-V3.1-Terminus | Gemini 3 Pro |
|---|---|---|
| LMArena Non-English | 1407 | 1474 |
| LMArena Russian | 1436 | 1493 |
| LMArena Chinese | — | 1523 |
| LMArena French | — | 1492 |
| LMArena German | — | 1515 |
| LMArena Japanese | — | 1510 |
| LMArena Korean | — | 1448 |
| LMArena Spanish | — | 1470 |
Instruction Following Gemini 3 Pro leads
DeepSeek-V3.1-Terminus: 74.0 (#106), Gemini 3 Pro: 76.3 (#45)
| Benchmark | DeepSeek-V3.1-Terminus | Gemini 3 Pro |
|---|---|---|
| LMArena Instruction Following | 1404 | 1458 |
| IFEval | — | 87.7% |
Long Context Too close to call
DeepSeek-V3.1-Terminus: 43.4 (#97), Gemini 3 Pro: 44.0 (#79)
| Benchmark | DeepSeek-V3.1-Terminus | Gemini 3 Pro |
|---|---|---|
| LMArena Longer Query | 1421 | 1471 |
| CL-bench | — | 15.8% |
Writing & Preference Gemini 3 Pro leads
DeepSeek-V3.1-Terminus: 61.0 (#92), Gemini 3 Pro: 66.4 (#35)
| Benchmark | DeepSeek-V3.1-Terminus | Gemini 3 Pro |
|---|---|---|
| LMArena Text | 1419 | 1479 |
| LMArena Creative Writing | 1403 | 1482 |
| LMArena Multi-Turn | 1411 | 1484 |
| EQ-Bench Creative Writing | — | 1525 |
| WildBench | — | 85.9% |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Gemini 3 Pro?
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 43.1 on the Noometry Index.
Is DeepSeek-V3.1-Terminus or Gemini 3 Pro better for coding?
Gemini 3 Pro scores higher on coding benchmarks: 51.6 versus 42.0 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1-Terminus and Gemini 3 Pro share?
13 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Gemini 3 Pro has 67.