# Gemma 3 4B vs Kimi K3

> Kimi K3 is the stronger model overall, scoring 59.5 to 28.1 on the Noometry Index. Gemma 3 4B costs 120× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemma-3-4b-vs-kimi-k3
- Last updated: 2026-10-10
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. Gemma 3 4B scores higher in 0 categories and Kimi K3 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K3 leads 74.2 to 16.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 7.5% for Gemma 3 4B and 97.2% for Kimi K3.
- Gemma 3 4B is cheaper at $0.04 / $0.08 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 accepts more context: 1.05M tokens versus 131K.

## Snapshot

| | Gemma 3 4B | Kimi K3 |
|---|---|---|
| Provider | Google | Moonshot AI |
| Noometry Index | 28.1 | 59.5 |
| Rank | 326 | 15 |
| Context | 131K | 1.05M |
| Input $/M | $0.04 | $3 |
| Output $/M | $0.08 | $15 |
| Weights | Open | Open |

## Coding

- Gemma 3 4B: 35.9 (#215)
- Kimi K3: 61.0 (#10)

| Benchmark | Gemma 3 4B | Kimi K3 |
|---|---|---|
| LMArena Coding | 1230 | 1508 |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| LMArena WebDev | — | 1654 |
| FrontierSWE | — | 25.9% |
| SciCode | — | 59.5% |
| WeirdML | — | 82.6% |
| ALE-Bench | — | 1,524 |

## Agentic & Tool Use

- Gemma 3 4B: 20.9 (#142)
- Kimi K3: 41.8 (#20)

| Benchmark | Gemma 3 4B | Kimi K3 |
|---|---|---|
| APEX-Agents | — | 50.6% |
| Berkeley Function Calling Leaderboard | 19.6% | — |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| Vending-Bench 2 | — | 5,165 |

## Reasoning

- Gemma 3 4B: 13.2 (#335)
- Kimi K3: 63.0 (#17)

| Benchmark | Gemma 3 4B | Kimi K3 |
|---|---|---|
| Chess Puzzles | 0% | 39% |
| LMArena Hard Prompts | 1253 | 1496 |
| DTBench | 50.9% | 91.2% |
| LMCA | 2.8% | 52.7% |
| Epoch Capabilities Index | 116.02 | 157.45 |
| ARC-AGI-2 | — | 60.4% |
| SimpleBench | — | 60.7% |
| Kagi LLM Benchmark | 25.2% | — |
| NYT Connections (extended) | — | 93.6% |
| ARC-AGI-1 | — | 94.5% |
| CritPt | — | 23.4% |
| Mystery Game Puzzles | — | 26% |
| Surface Evolver Bench | — | 95% |
| ForecastBench | — | 61.1 |

## Math

- Gemma 3 4B: 16.8 (#292)
- Kimi K3: 74.2 (#16)

| Benchmark | Gemma 3 4B | Kimi K3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 7.5% | 97.2% |
| LMArena Math | 1239 | 1491 |
| FrontierMath (Tiers 1-3) | — | 72.2% |
| FrontierMath Tier 4 | — | 39% |
| MathArena Final-Answer Competitions | — | 87.8% |
| ProofBench | — | 87% |

## Knowledge

- Gemma 3 4B: 11.8 (#299)
- Kimi K3: 63.2 (#21)

| Benchmark | Gemma 3 4B | Kimi K3 |
|---|---|---|
| GPQA Diamond | 23.2% | 93.1% |
| LMArena Expert | 1223 | 1521 |
| SimpleQA Verified | — | 50.6% |
| Vectara Hallucination Rate | 6.4% | — |

## Multimodal

- Gemma 3 4B: —
- Kimi K3: 37.8 (#70)

| Benchmark | Gemma 3 4B | Kimi K3 |
|---|---|---|
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |

## Multilingual

- Gemma 3 4B: 42.5 (#194)
- Kimi K3: 56.3 (#21)

| Benchmark | Gemma 3 4B | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1273 | 1466 |
| LMArena German | 1281 | 1488 |
| LMArena Russian | 1294 | 1482 |
| LMArena Chinese | — | 1529 |
| LMArena French | — | 1491 |
| LMArena Japanese | — | 1487 |
| LMArena Korean | — | 1458 |
| LMArena Spanish | — | 1472 |

## Instruction Following

- Gemma 3 4B: 65.2 (#225)
- Kimi K3: 77.7 (#14)

| Benchmark | Gemma 3 4B | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1239 | 1483 |

## Long Context

- Gemma 3 4B: 38.7 (#194)
- Kimi K3: 45.8 (#29)

| Benchmark | Gemma 3 4B | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1273 | 1494 |

## Writing & Preference

- Gemma 3 4B: 42.0 (#239)
- Kimi K3: 76.6 (#4)

| Benchmark | Gemma 3 4B | Kimi K3 |
|---|---|---|
| LMArena Text | 1291 | 1476 |
| LMArena Creative Writing | 1271 | 1454 |
| EQ-Bench Creative Writing | 1068 | 2082 |
| LMArena Multi-Turn | 1255 | 1488 |
| EQ-Bench 4 | — | 1339 |

## FAQ

### Is Gemma 3 4B better than Kimi K3?

Kimi K3 is the stronger model overall, scoring 59.5 to 28.1 on the Noometry Index. Gemma 3 4B costs 120× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.

### Which is cheaper, Gemma 3 4B or Kimi K3?

Gemma 3 4B is cheaper. It lists at $0.04 per million input tokens and $0.08 per million output tokens; Kimi K3 lists at $3 and $15.

### Is Gemma 3 4B or Kimi K3 better for coding?

Kimi K3 scores higher on coding benchmarks: 61.0 versus 35.9 in the Noometry coding category.

### Which has the bigger context window?

Kimi K3 does, with 1.05M tokens against 131K.

### How many benchmarks do Gemma 3 4B and Kimi K3 share?

19 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and Kimi K3 has 53.