# Gemini 3.7 Flash vs Kimi K3

> Gemini 3.7 Flash and Kimi K3 score almost the same on the Noometry Index (59.8 vs 59.5), so choose on price, context window or the category you care about most.

- Canonical page: https://noometry.com/compare/gemini-3-7-flash-vs-kimi-k3
- Last updated: 2026-10-11
- Shared benchmarks: 42

## Summary

- They share 42 benchmarks with published results for both. Gemini 3.7 Flash scores higher in 4 categories and Kimi K3 in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.7 Flash leads 70.0 to 63.0.
- The biggest single-benchmark swing is ProofBench: 58% for Gemini 3.7 Flash and 87% for Kimi K3.
- Gemini 3.7 Flash is cheaper at $0.75 / $3.75 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 has downloadable open weights; the other is API-only.

## Snapshot

| | Gemini 3.7 Flash | Kimi K3 |
|---|---|---|
| Provider | Google | Moonshot AI |
| Noometry Index | 59.8 | 59.5 |
| Rank | 14 | 15 |
| Context | 1.05M | 1.05M |
| Input $/M | $0.75 | $3 |
| Output $/M | $3.75 | $15 |
| Weights | Proprietary | Open |

## Coding

- Gemini 3.7 Flash: 56.2 (#22)
- Kimi K3: 61.0 (#10)

| Benchmark | Gemini 3.7 Flash | Kimi K3 |
|---|---|---|
| DeepSWE | 65.5% | 68.5% |
| FrontierCode | 43.6% | 44.2% |
| LMArena WebDev | 1592 | 1654 |
| FrontierSWE | 20.3% | 25.9% |
| SciCode | 59.8% | 59.5% |
| LMArena Coding | 1497 | 1508 |
| ALE-Bench | 904.3 | 1,524 |
| WeirdML | — | 82.6% |

## Agentic & Tool Use

- Gemini 3.7 Flash: 42.1 (#19)
- Kimi K3: 41.8 (#20)

| Benchmark | Gemini 3.7 Flash | Kimi K3 |
|---|---|---|
| APEX-Agents | 67.8% | 50.6% |
| GDP.pdf | 23.8% | 19% |
| Remote Labor Index | 5% | — |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| Vending-Bench 2 | — | 5,165 |

## Reasoning

- Gemini 3.7 Flash: 70.0 (#15)
- Kimi K3: 63.0 (#17)

| Benchmark | Gemini 3.7 Flash | Kimi K3 |
|---|---|---|
| ARC-AGI-2 | 84.6% | 60.4% |
| NYT Connections (extended) | 94% | 93.6% |
| ARC-AGI-1 | 95.5% | 94.5% |
| CritPt | 14.3% | 23.4% |
| Chess Puzzles | 47% | 39% |
| LMArena Hard Prompts | 1494 | 1496 |
| Mystery Game Puzzles | 37% | 26% |
| DTBench | 96.8% | 91.2% |
| LMCA | 50.4% | 52.7% |
| Epoch Capabilities Index | 157.27 | 157.45 |
| SimpleBench | — | 60.7% |
| Surface Evolver Bench | — | 95% |
| ForecastBench | — | 61.1 |

## Math

- Gemini 3.7 Flash: 69.6 (#23)
- Kimi K3: 74.2 (#16)

| Benchmark | Gemini 3.7 Flash | Kimi K3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 71.6% | 72.2% |
| FrontierMath Tier 4 | 36.6% | 39% |
| OTIS Mock AIME 2024-2025 | 97.2% | 97.2% |
| ProofBench | 58% | 87% |
| LMArena Math | 1507 | 1491 |
| MathArena Final-Answer Competitions | — | 87.8% |

## Knowledge

- Gemini 3.7 Flash: 69.7 (#5)
- Kimi K3: 63.2 (#21)

| Benchmark | Gemini 3.7 Flash | Kimi K3 |
|---|---|---|
| GPQA Diamond | 94.8% | 93.1% |
| SimpleQA Verified | 69.2% | 50.6% |
| LMArena Expert | 1508 | 1521 |

## Multimodal

- Gemini 3.7 Flash: 37.3 (#73)
- Kimi K3: 37.8 (#70)

| Benchmark | Gemini 3.7 Flash | Kimi K3 |
|---|---|---|
| Furniture Assembly | 26.7% | 34.2% |
| LMArena Vision | 1316 | — |
| Blueprint-Bench 2 | — | 29.5% |

## Multilingual

- Gemini 3.7 Flash: 57.6 (#7)
- Kimi K3: 56.3 (#21)

| Benchmark | Gemini 3.7 Flash | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1484 | 1466 |
| LMArena Chinese | 1548 | 1529 |
| LMArena French | 1505 | 1491 |
| LMArena German | 1498 | 1488 |
| LMArena Japanese | 1512 | 1487 |
| LMArena Korean | 1483 | 1458 |
| LMArena Russian | 1516 | 1482 |
| LMArena Spanish | 1503 | 1472 |

## Instruction Following

- Gemini 3.7 Flash: 77.7 (#15)
- Kimi K3: 77.7 (#14)

| Benchmark | Gemini 3.7 Flash | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1483 | 1483 |

## Long Context

- Gemini 3.7 Flash: 45.7 (#30)
- Kimi K3: 45.8 (#29)

| Benchmark | Gemini 3.7 Flash | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1492 | 1494 |

## Writing & Preference

- Gemini 3.7 Flash: 71.2 (#20)
- Kimi K3: 76.6 (#4)

| Benchmark | Gemini 3.7 Flash | Kimi K3 |
|---|---|---|
| LMArena Text | 1486 | 1476 |
| LMArena Creative Writing | 1490 | 1454 |
| EQ-Bench Creative Writing | 1723 | 2082 |
| LMArena Multi-Turn | 1489 | 1488 |
| EQ-Bench 4 | — | 1339 |

## FAQ

### Is Gemini 3.7 Flash better than Kimi K3?

Gemini 3.7 Flash and Kimi K3 score almost the same on the Noometry Index (59.8 vs 59.5), so choose on price, context window or the category you care about most.

### Which is cheaper, Gemini 3.7 Flash or Kimi K3?

Gemini 3.7 Flash is cheaper. It lists at $0.75 per million input tokens and $3.75 per million output tokens; Kimi K3 lists at $3 and $15.

### Is Gemini 3.7 Flash or Kimi K3 better for coding?

Kimi K3 scores higher on coding benchmarks: 61.0 versus 56.2 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1.05M tokens.

### How many benchmarks do Gemini 3.7 Flash and Kimi K3 share?

42 benchmarks have published results for both models. Gemini 3.7 Flash has 44 scored results on Noometry and Kimi K3 has 53.