# Gemini 3.1 Pro Preview vs Kimi K2.6

> Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 47.5 on the Noometry Index. Kimi K2.6 costs 2.6× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemini-3-1-pro-preview-vs-kimi-k2-6
- Last updated: 2026-10-10
- Shared benchmarks: 50

## Summary

- They share 50 benchmarks with published results for both. Gemini 3.1 Pro Preview scores higher in 8 categories and Kimi K2.6 in 2 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 39.6.
- The biggest single-benchmark swing is SimpleQA Verified: 73.5% for Gemini 3.1 Pro Preview and 34.9% for Kimi K2.6.
- Kimi K2.6 is cheaper at $0.95 / $4 per million input/output tokens, against $2 / $12 for Gemini 3.1 Pro Preview.
- Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 262K.
- Kimi K2.6 has downloadable open weights; the other is API-only.

## Snapshot

| | Gemini 3.1 Pro Preview | Kimi K2.6 |
|---|---|---|
| Provider | Google | Moonshot AI |
| Noometry Index | 56.7 | 47.5 |
| Rank | 20 | 52 |
| Context | 1.05M | 262K |
| Input $/M | $2 | $0.95 |
| Output $/M | $12 | $4 |
| Weights | Proprietary | Open |

## Coding

- Gemini 3.1 Pro Preview: 42.7 (#88)
- Kimi K2.6: 50.5 (#41)

| Benchmark | Gemini 3.1 Pro Preview | Kimi K2.6 |
|---|---|---|
| SWE-bench Verified | 75.6% | 76.7% |
| LMArena WebDev | 1446 | 1509 |
| SciCode | 58.9% | 53.5% |
| WeirdML | 72.1% | 55.9% |
| LMArena Coding | 1483 | 1488 |
| ALE-Bench | 1,161 | 1,093 |
| DeepSWE | 11.7% | — |
| GSO | 22.6% | — |
| MirrorCode | 8.9% | — |
| AlgoTune | 2.02 | — |

## Agentic & Tool Use

- Gemini 3.1 Pro Preview: 37.7 (#33)
- Kimi K2.6: 21.8 (#128)

| Benchmark | Gemini 3.1 Pro Preview | Kimi K2.6 |
|---|---|---|
| ExploitBench | 26.1% | 18.4% |
| GBAEval | 0.8% | 0.9% |
| GDP.pdf | 17% | 12% |
| Vending-Bench 2 | 3,774 | 6,205 |
| Terminal-Bench | 80.2% | — |
| APEX-Agents | 35.3% | — |
| OSWorld 2.0 | — | 4.6% |
| τ²-bench Banking | 26% | — |
| DeepResearch Bench | 47.8% | — |
| PostTrainBench | 22% | — |
| BALROG | 57% | — |
| LMArena Search | 1211 | — |
| METR Time Horizons | 77% | — |

## Reasoning

- Gemini 3.1 Pro Preview: 71.7 (#9)
- Kimi K2.6: 39.6 (#51)

| Benchmark | Gemini 3.1 Pro Preview | Kimi K2.6 |
|---|---|---|
| NYT Connections (extended) | 97.4% | 87.2% |
| CritPt | 17.7% | 8% |
| Chess Puzzles | 55% | 26% |
| EBR-Bench | 14.3% | 2.4% |
| LMArena Hard Prompts | 1484 | 1470 |
| Mystery Game Puzzles | 34% | 18% |
| DTBench | 97.1% | 90.9% |
| LMCA | 53.8% | 37.3% |
| Epoch Capabilities Index | 154.92 | 151.03 |
| ARC-AGI-2 | 77.1% | — |
| SimpleBench | 79.6% | — |
| ARC-AGI-1 | 98% | — |
| EnigmaEval | 36.8% | — |
| Thematic Generalization | 79.4% | — |
| ForecastBench | 59 | — |

## Math

- Gemini 3.1 Pro Preview: 61.8 (#28)
- Kimi K2.6: 56.8 (#34)

| Benchmark | Gemini 3.1 Pro Preview | Kimi K2.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 59.6% | 57.2% |
| FrontierMath Tier 4 | 26.8% | 25.6% |
| MathArena Final-Answer Competitions | 86.5% | 72.9% |
| OTIS Mock AIME 2024-2025 | 95.6% | 96.1% |
| ProofBench | 26% | 16% |
| LMArena Math | 1485 | 1475 |
| FrontierMath (Feb 2025 set) | 36.9% | 39% |
| FrontierMath Tier 4 (v1) | 16.7% | 14.6% |

## Knowledge

- Gemini 3.1 Pro Preview: 71.8 (#3)
- Kimi K2.6: 53.9 (#48)

| Benchmark | Gemini 3.1 Pro Preview | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 94.4% | 90.8% |
| SimpleQA Verified | 73.5% | 34.9% |
| Vectara Hallucination Rate | 10.4% | 10.8% |
| LMArena Expert | 1484 | 1491 |
| Humanity's Last Exam | 46.4% | — |

## Multimodal

- Gemini 3.1 Pro Preview: 38.1 (#60)
- Kimi K2.6: 31.8 (#95)

| Benchmark | Gemini 3.1 Pro Preview | Kimi K2.6 |
|---|---|---|
| LMArena Vision | 1296 | 1282 |
| Blueprint-Bench 2 | 26.5% | 3.9% |
| Furniture Assembly | 26.7% | 21.7% |
| LMArena Document | 1444 | 1451 |

## Multilingual

- Gemini 3.1 Pro Preview: 57.1 (#11)
- Kimi K2.6: 55.1 (#35)

| Benchmark | Gemini 3.1 Pro Preview | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | 1475 | 1446 |
| LMArena Chinese | 1528 | 1521 |
| LMArena French | 1488 | 1471 |
| LMArena German | 1491 | 1449 |
| LMArena Japanese | 1494 | 1442 |
| LMArena Korean | 1458 | 1428 |
| LMArena Russian | 1496 | 1446 |
| LMArena Spanish | 1480 | 1463 |

## Instruction Following

- Gemini 3.1 Pro Preview: 77.1 (#31)
- Kimi K2.6: 76.5 (#40)

| Benchmark | Gemini 3.1 Pro Preview | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | 1465 | 1451 |

## Long Context

- Gemini 3.1 Pro Preview: 47.3 (#17)
- Kimi K2.6: 44.9 (#50)

| Benchmark | Gemini 3.1 Pro Preview | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | 1482 | 1468 |
| CL-bench | 20.8% | — |
| CL-bench Life | 16.9% | — |

## Writing & Preference

- Gemini 3.1 Pro Preview: 66.3 (#36)
- Kimi K2.6: 68.7 (#25)

| Benchmark | Gemini 3.1 Pro Preview | Kimi K2.6 |
|---|---|---|
| LMArena Text | 1480 | 1455 |
| LMArena Creative Writing | 1482 | 1434 |
| EQ-Bench Creative Writing | 1491 | 1725 |
| EQ-Bench 4 | 1142 | 1202 |
| LMArena Multi-Turn | 1486 | 1453 |

## FAQ

### Is Gemini 3.1 Pro Preview better than Kimi K2.6?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 47.5 on the Noometry Index. Kimi K2.6 costs 2.6× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.

### Which is cheaper, Gemini 3.1 Pro Preview or Kimi K2.6?

Kimi K2.6 is cheaper. It lists at $0.95 per million input tokens and $4 per million output tokens; Gemini 3.1 Pro Preview lists at $2 and $12.

### Is Gemini 3.1 Pro Preview or Kimi K2.6 better for coding?

Kimi K2.6 scores higher on coding benchmarks: 50.5 versus 42.7 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.1 Pro Preview does, with 1.05M tokens against 262K.

### How many benchmarks do Gemini 3.1 Pro Preview and Kimi K2.6 share?

50 benchmarks have published results for both models. Gemini 3.1 Pro Preview has 71 scored results on Noometry and Kimi K2.6 has 51.