# Kimi K3 vs Llama 3.1-405B

> Kimi K3 is the stronger model overall, scoring 59.5 to 30.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/kimi-k3-vs-llama-3-1-405b
- Last updated: 2026-10-10
- Shared benchmarks: 25

## Summary

- They share 25 benchmarks with published results for both. Kimi K3 scores higher in 9 categories and Llama 3.1-405B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K3 leads 74.0 to 17.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 97.2% for Kimi K3 and 9.7% for Llama 3.1-405B.

## Snapshot

| | Kimi K3 | Llama 3.1-405B |
|---|---|---|
| Provider | Moonshot AI | Meta |
| Noometry Index | 59.5 | 30.4 |
| Rank | 12 | 279 |
| Context | 1.05M | — |
| Input $/M | $3 | — |
| Output $/M | $15 | — |
| Weights | Open | Open |

## Coding

- Kimi K3: 60.6 (#8)
- Llama 3.1-405B: 32.7 (#257)

| Benchmark | Kimi K3 | Llama 3.1-405B |
|---|---|---|
| WeirdML | 82.6% | 21.4% |
| LMArena Coding | 1509 | 1291 |
| DeepSWE | 68.5% | — |
| FrontierCode | 44.2% | — |
| LMArena WebDev | 1659 | — |
| FrontierSWE | 25.9% | — |
| SciCode | 59.5% | — |
| ALE-Bench | 1,524 | — |

## Agentic & Tool Use

- Kimi K3: 41.8 (#19)
- Llama 3.1-405B: 20.9 (#131)

| Benchmark | Kimi K3 | Llama 3.1-405B |
|---|---|---|
| APEX-Agents | 50.6% | — |
| TheAgentCompany | — | 7.4% |
| τ²-bench Banking | 37.1% | — |
| Cybench | — | 7.5% |
| PostTrainBench | 32% | — |
| GBAEval | 48.3% | — |
| GDP.pdf | 19% | — |
| Vending-Bench 2 | 5,165 | — |

## Reasoning

- Kimi K3: 63.0 (#15)
- Llama 3.1-405B: 16.4 (#293)

| Benchmark | Kimi K3 | Llama 3.1-405B |
|---|---|---|
| SimpleBench | 60.7% | 23% |
| LMArena Hard Prompts | 1495 | 1270 |
| DTBench | 91.2% | 61.4% |
| Epoch Capabilities Index | 157.68 | 128.76 |
| ForecastBench | 61.1 | 59.9 |
| ARC-AGI-2 | 60.4% | — |
| Kagi LLM Benchmark | — | 45% |
| NYT Connections (extended) | 93.6% | — |
| ARC-AGI-1 | 94.5% | — |
| CritPt | 23.4% | — |
| Chess Puzzles | 39% | — |
| Mystery Game Puzzles | 26% | — |
| LMCA | 52.7% | — |
| Surface Evolver Bench | 95% | — |
| BIG-Bench Hard | — | 82.9% |
| HellaSwag | — | 89.2% |
| PIQA | — | 85.9% |
| WinoGrande | — | 89.2% |

## Math

- Kimi K3: 74.0 (#13)
- Llama 3.1-405B: 17.3 (#283)

| Benchmark | Kimi K3 | Llama 3.1-405B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 97.2% | 9.7% |
| LMArena Math | 1493 | 1281 |
| FrontierMath (Tiers 1-3) | 72.2% | — |
| FrontierMath Tier 4 | 39% | — |
| MathArena Final-Answer Competitions | 87.8% | — |
| ProofBench | 87% | — |
| Omni-MATH | — | 24.9% |
| MATH Level 5 | — | 49.8% |

## Knowledge

- Kimi K3: 63.2 (#19)
- Llama 3.1-405B: 30.4 (#221)

| Benchmark | Kimi K3 | Llama 3.1-405B |
|---|---|---|
| GPQA Diamond | 93.1% | 50.9% |
| LMArena Expert | 1520 | 1243 |
| SimpleQA Verified | 50.6% | — |
| MMLU-Pro | — | 72.3% |
| Confabulations | — | 17.6% |
| GPQA (HELM) | — | 52.2% |
| ARC (AI2) Challenge | — | 95.3% |
| MMLU | — | 84.5% |
| TriviaQA | — | 82.7% |

## Multimodal

- Kimi K3: 38.4 (#58)
- Llama 3.1-405B: —

| Benchmark | Kimi K3 | Llama 3.1-405B |
|---|---|---|
| Blueprint-Bench 2 | 29.5% | — |
| Furniture Assembly | 34.2% | — |

## Multilingual

- Kimi K3: 56.5 (#20)
- Llama 3.1-405B: 40.9 (#209)

| Benchmark | Kimi K3 | Llama 3.1-405B |
|---|---|---|
| LMArena Non-English | 1466 | 1248 |
| LMArena Chinese | 1528 | 1242 |
| LMArena French | 1488 | 1278 |
| LMArena German | 1474 | 1252 |
| LMArena Japanese | 1492 | 1208 |
| LMArena Korean | 1465 | 1184 |
| LMArena Russian | 1481 | 1265 |
| LMArena Spanish | 1478 | 1259 |

## Instruction Following

- Kimi K3: 77.8 (#13)
- Llama 3.1-405B: 66.1 (#209)

| Benchmark | Kimi K3 | Llama 3.1-405B |
|---|---|---|
| LMArena Instruction Following | 1481 | 1259 |
| IFEval | — | 81.1% |

## Long Context

- Kimi K3: 45.8 (#27)
- Llama 3.1-405B: 38.4 (#192)

| Benchmark | Kimi K3 | Llama 3.1-405B |
|---|---|---|
| LMArena Longer Query | 1494 | 1266 |

## Writing & Preference

- Kimi K3: 76.7 (#4)
- Llama 3.1-405B: 39.1 (#246)

| Benchmark | Kimi K3 | Llama 3.1-405B |
|---|---|---|
| LMArena Text | 1476 | 1284 |
| LMArena Creative Writing | 1452 | 1262 |
| EQ-Bench Creative Writing | 2082 | 870 |
| LMArena Multi-Turn | 1488 | 1297 |
| WildBench | — | 78.3% |
| EQ-Bench 4 | 1339 | — |

## FAQ

### Is Kimi K3 better than Llama 3.1-405B?

Kimi K3 is the stronger model overall, scoring 59.5 to 30.4 on the Noometry Index.

### Is Kimi K3 or Llama 3.1-405B better for coding?

Kimi K3 scores higher on coding benchmarks: 60.6 versus 32.7 in the Noometry coding category.

### How many benchmarks do Kimi K3 and Llama 3.1-405B share?

25 benchmarks have published results for both models. Kimi K3 has 53 scored results on Noometry and Llama 3.1-405B has 42.