Model comparison
Gemini 3 Pro vs Kimi K2.6
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 47.7 on the Noometry Index.
Last verified . 36 shared benchmarks.
GoogleSummary
- They share 36 benchmarks with published results for both. Gemini 3 Pro scores higher in 6 categories and Kimi K2.6 in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in multimodal, where Gemini 3 Pro leads 57.6 to 31.6.
- The biggest single-benchmark swing is WeirdML: 69.9% for Gemini 3 Pro and 55.9% for Kimi K2.6.
- Kimi K2.6 has downloadable open weights; the other is API-only.
Side by side
| Gemini 3 Pro | Kimi K2.6 | |
|---|---|---|
| Provider | Moonshot AI | |
| Noometry Index | 54.8 | 47.7 |
| Released | 2025-11-18 | 2026-04-20 |
| Weights | Proprietary | Open |
| Context window | — | 262K |
| Max output | — | 262K |
| Input $ / M tokens | — | $0.95 |
| Output $ / M tokens | — | $4 |
| Results tracked | 67 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemini 3 Pro: 51.6 (#39), Kimi K2.6: 50.7 (#43)
| Benchmark | Gemini 3 Pro | Kimi K2.6 |
|---|---|---|
| SWE-bench Verified | 72.9% | 76.7% |
| LMArena WebDev | 1440 | 1509 |
| WeirdML | 69.9% | 55.9% |
| LMArena Coding | 1481 | 1488 |
| ALE-Bench | 1,177 | 1,093 |
| SWE-bench Verified (bash only) | 74.2% | — |
| SWE-bench Multilingual | 68.7% | — |
| SciCode | — | 53.5% |
| GSO | 18.6% | — |
| AlgoTune | 1.83 | — |
Agentic & Tool Use Gemini 3 Pro leads
Gemini 3 Pro: 40.6 (#23), Kimi K2.6: 21.9 (#137)
| Benchmark | Gemini 3 Pro | Kimi K2.6 |
|---|---|---|
| Vending-Bench 2 | 5,478 | 6,205 |
| Terminal-Bench | 69.4% | — |
| Berkeley Function Calling Leaderboard | 72.5% | — |
| OSWorld 2.0 | — | 4.6% |
| 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% | — |
| ExploitBench | — | 18.4% |
| GBAEval | — | 0.9% |
| GDP.pdf | — | 12% |
| LMArena Search | 1207 | — |
| METR Time Horizons | 71% | — |
Reasoning Gemini 3 Pro leads
Gemini 3 Pro: 52.5 (#31), Kimi K2.6: 40.5 (#55)
| Benchmark | Gemini 3 Pro | Kimi K2.6 |
|---|---|---|
| NYT Connections (extended) | 94.4% | 87.2% |
| CritPt | 6.9% | 8% |
| Chess Puzzles | 31% | 26% |
| LMArena Hard Prompts | 1480 | 1470 |
| Epoch Capabilities Index | 152.92 | 151.05 |
| ARC-AGI-2 | 31.1% | — |
| SimpleBench | 76.4% | — |
| Kagi LLM Benchmark | 80.1% | — |
| ARC-AGI-1 | 75% | — |
| EnigmaEval | 18.2% | — |
| EBR-Bench | — | 2.4% |
| Mystery Game Puzzles | — | 18% |
| DTBench | — | 90.9% |
| LMCA | — | 37.3% |
| ForecastBench | 61.2 | — |
Math Kimi K2.6 leads
Gemini 3 Pro: 49.9 (#59), Kimi K2.6: 57.0 (#41)
| Benchmark | Gemini 3 Pro | Kimi K2.6 |
|---|---|---|
| MathArena Final-Answer Competitions | 67% | 72.9% |
| OTIS Mock AIME 2024-2025 | 91.4% | 96.1% |
| ProofBench | 20% | 16% |
| LMArena Math | 1476 | 1475 |
| FrontierMath (Feb 2025 set) | 37.6% | 39% |
| FrontierMath Tier 4 (v1) | 18.8% | 14.6% |
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 25.6% |
| Omni-MATH | 55.5% | — |
Knowledge Gemini 3 Pro leads
Gemini 3 Pro: 64.4 (#16), Kimi K2.6: 54.0 (#54)
| Benchmark | Gemini 3 Pro | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 92.6% | 90.8% |
| Vectara Hallucination Rate | 13.6% | 10.8% |
| LMArena Expert | 1475 | 1491 |
| Humanity's Last Exam | 37.5% | — |
| SimpleQA Verified | — | 34.9% |
| MMLU-Pro | 90.3% | — |
| GPQA (HELM) | 80.3% | — |
Multimodal Gemini 3 Pro leads
Gemini 3 Pro: 57.6 (#2), Kimi K2.6: 31.6 (#103)
| Benchmark | Gemini 3 Pro | Kimi K2.6 |
|---|---|---|
| LMArena Vision | 1305 | 1283 |
| LMArena Document | 1434 | 1451 |
| GeoBench | 84% | — |
| VPCT | 91% | — |
| Blueprint-Bench 2 | — | 3.9% |
| Furniture Assembly | — | 21.7% |
Multilingual Gemini 3 Pro leads
Gemini 3 Pro: 56.9 (#16), Kimi K2.6: 54.9 (#37)
| Benchmark | Gemini 3 Pro | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | 1474 | 1446 |
| LMArena Chinese | 1523 | 1521 |
| LMArena French | 1492 | 1471 |
| LMArena German | 1515 | 1450 |
| LMArena Japanese | 1510 | 1443 |
| LMArena Korean | 1448 | 1427 |
| LMArena Russian | 1493 | 1446 |
| LMArena Spanish | 1470 | 1464 |
Instruction Following Too close to call
Gemini 3 Pro: 76.3 (#45), Kimi K2.6: 76.3 (#43)
| Benchmark | Gemini 3 Pro | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | 1458 | 1451 |
| IFEval | 87.7% | — |
Long Context Too close to call
Gemini 3 Pro: 44.0 (#79), Kimi K2.6: 44.9 (#52)
| Benchmark | Gemini 3 Pro | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | 1471 | 1468 |
| CL-bench | 15.8% | — |
Writing & Preference Kimi K2.6 leads
Gemini 3 Pro: 66.4 (#35), Kimi K2.6: 68.5 (#26)
| Benchmark | Gemini 3 Pro | Kimi K2.6 |
|---|---|---|
| LMArena Text | 1479 | 1455 |
| LMArena Creative Writing | 1482 | 1434 |
| EQ-Bench Creative Writing | 1525 | 1725 |
| LMArena Multi-Turn | 1484 | 1453 |
| WildBench | 85.9% | — |
| EQ-Bench 4 | — | 1202 |
Frequently asked questions
Is Gemini 3 Pro better than Kimi K2.6?
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 47.7 on the Noometry Index.
Is Gemini 3 Pro or Kimi K2.6 better for coding?
They score almost the same on coding (51.6 vs 50.7); test both on your own repository before choosing.
How many benchmarks do Gemini 3 Pro and Kimi K2.6 share?
36 benchmarks have published results for both models. Gemini 3 Pro has 67 scored results on Noometry and Kimi K2.6 has 51.