# Claude 3 Haiku vs Kimi K3

> Kimi K3 is the stronger model overall, scoring 59.5 to 25.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-kimi-k3
- Last updated: 2026-10-10
- Shared benchmarks: 25

## Summary

- They share 25 benchmarks with published results for both. Claude 3 Haiku 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 74.0 to 9.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 97.2% for Kimi K3.
- Kimi K3 has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 3 Haiku | Kimi K3 |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 25.7 | 59.5 |
| Rank | 333 | 12 |
| Context | — | 1.05M |
| Input $/M | — | $3 |
| Output $/M | — | $15 |
| Weights | Proprietary | Open |

## Coding

- Claude 3 Haiku: 26.1 (#320)
- Kimi K3: 60.6 (#8)

| Benchmark | Claude 3 Haiku | Kimi K3 |
|---|---|---|
| WeirdML | 9.8% | 82.6% |
| LMArena Coding | 1199 | 1509 |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| LMArena WebDev | — | 1659 |
| FrontierSWE | — | 25.9% |
| SciCode | — | 59.5% |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| ALE-Bench | — | 1,524 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Agentic & Tool Use

- Claude 3 Haiku: —
- Kimi K3: 41.8 (#19)

| Benchmark | Claude 3 Haiku | Kimi K3 |
|---|---|---|
| APEX-Agents | — | 50.6% |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| Vending-Bench 2 | — | 5,165 |

## Reasoning

- Claude 3 Haiku: 15.7 (#303)
- Kimi K3: 63.0 (#15)

| Benchmark | Claude 3 Haiku | Kimi K3 |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1495 |
| DTBench | 50.1% | 91.2% |
| LMCA | 8.8% | 52.7% |
| Epoch Capabilities Index | 118.3 | 157.68 |
| ForecastBench | 53.2 | 61.1 |
| ARC-AGI-2 | — | 60.4% |
| SimpleBench | — | 60.7% |
| Kagi LLM Benchmark | 34.2% | — |
| NYT Connections (extended) | — | 93.6% |
| ARC-AGI-1 | — | 94.5% |
| CritPt | — | 23.4% |
| Chess Puzzles | — | 39% |
| Mystery Game Puzzles | — | 26% |
| Surface Evolver Bench | — | 95% |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.2 (#313)
- Kimi K3: 74.0 (#13)

| Benchmark | Claude 3 Haiku | Kimi K3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 97.2% |
| LMArena Math | 1188 | 1493 |
| FrontierMath (Tiers 1-3) | — | 72.2% |
| FrontierMath Tier 4 | — | 39% |
| MathArena Final-Answer Competitions | — | 87.8% |
| ProofBench | — | 87% |
| MATH Level 5 | 14.9% | — |

## Knowledge

- Claude 3 Haiku: 17.3 (#280)
- Kimi K3: 63.2 (#19)

| Benchmark | Claude 3 Haiku | Kimi K3 |
|---|---|---|
| GPQA Diamond | 36.3% | 93.1% |
| LMArena Expert | 1148 | 1520 |
| SimpleQA Verified | — | 50.6% |
| Confabulations | 34.2% | — |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.1 (#123)
- Kimi K3: 38.4 (#58)

| Benchmark | Claude 3 Haiku | Kimi K3 |
|---|---|---|
| LMArena Vision | 950 | — |
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.1 (#238)
- Kimi K3: 56.5 (#20)

| Benchmark | Claude 3 Haiku | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1178 | 1466 |
| LMArena Chinese | 1155 | 1528 |
| LMArena French | 1194 | 1488 |
| LMArena German | 1174 | 1474 |
| LMArena Japanese | 1102 | 1492 |
| LMArena Korean | 1108 | 1465 |
| LMArena Russian | 1204 | 1481 |
| LMArena Spanish | 1165 | 1478 |

## Instruction Following

- Claude 3 Haiku: 61.5 (#242)
- Kimi K3: 77.8 (#13)

| Benchmark | Claude 3 Haiku | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1174 | 1481 |

## Long Context

- Claude 3 Haiku: 36.1 (#232)
- Kimi K3: 45.8 (#27)

| Benchmark | Claude 3 Haiku | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1191 | 1494 |

## Writing & Preference

- Claude 3 Haiku: 29.8 (#286)
- Kimi K3: 76.7 (#4)

| Benchmark | Claude 3 Haiku | Kimi K3 |
|---|---|---|
| LMArena Text | 1195 | 1476 |
| LMArena Creative Writing | 1157 | 1452 |
| EQ-Bench Creative Writing | 717 | 2082 |
| LMArena Multi-Turn | 1190 | 1488 |
| EQ-Bench 4 | — | 1339 |

## FAQ

### Is Claude 3 Haiku better than Kimi K3?

Kimi K3 is the stronger model overall, scoring 59.5 to 25.7 on the Noometry Index.

### Is Claude 3 Haiku or Kimi K3 better for coding?

Kimi K3 scores higher on coding benchmarks: 60.6 versus 26.1 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and Kimi K3 share?

25 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Kimi K3 has 53.