Model comparison
Claude 3 Haiku vs Kimi K2.6
Kimi K2.6 is the stronger model overall, scoring 47.5 to 25.7 on the Noometry Index.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Kimi K2.6 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.6 leads 56.8 to 9.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 96.1% for Kimi K2.6.
- Kimi K2.6 has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Haiku | Kimi K2.6 | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 25.7 | 47.5 |
| Released | 2024-03-07 | 2026-04-20 |
| Weights | Proprietary | Open |
| Context window | — | 262K |
| Max output | — | 262K |
| Input $ / M tokens | — | $0.95 |
| Output $ / M tokens | — | $4 |
| Results tracked | 37 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.6 leads
Claude 3 Haiku: 26.1 (#320), Kimi K2.6: 50.5 (#41)
| Benchmark | Claude 3 Haiku | Kimi K2.6 |
|---|---|---|
| WeirdML | 9.8% | 55.9% |
| LMArena Coding | 1199 | 1488 |
| SWE-bench Verified | — | 76.7% |
| LMArena WebDev | — | 1509 |
| SciCode | — | 53.5% |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| ALE-Bench | — | 1,093 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |
Agentic & Tool Use Not comparable
Claude 3 Haiku: —, Kimi K2.6: 21.8 (#128)
| Benchmark | Claude 3 Haiku | Kimi K2.6 |
|---|---|---|
| OSWorld 2.0 | — | 4.6% |
| ExploitBench | — | 18.4% |
| GBAEval | — | 0.9% |
| GDP.pdf | — | 12% |
| Vending-Bench 2 | — | 6,205 |
Reasoning Kimi K2.6 leads
Claude 3 Haiku: 15.7 (#303), Kimi K2.6: 39.6 (#51)
| Benchmark | Claude 3 Haiku | Kimi K2.6 |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1470 |
| DTBench | 50.1% | 90.9% |
| LMCA | 8.8% | 37.3% |
| Epoch Capabilities Index | 118.3 | 151.03 |
| Kagi LLM Benchmark | 34.2% | — |
| NYT Connections (extended) | — | 87.2% |
| CritPt | — | 8% |
| Chess Puzzles | — | 26% |
| EBR-Bench | — | 2.4% |
| Mystery Game Puzzles | — | 18% |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |
Math Kimi K2.6 leads
Claude 3 Haiku: 9.2 (#313), Kimi K2.6: 56.8 (#34)
| Benchmark | Claude 3 Haiku | Kimi K2.6 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 96.1% |
| LMArena Math | 1188 | 1475 |
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 25.6% |
| MathArena Final-Answer Competitions | — | 72.9% |
| ProofBench | — | 16% |
| MATH Level 5 | 14.9% | — |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge Kimi K2.6 leads
Claude 3 Haiku: 17.3 (#280), Kimi K2.6: 53.9 (#48)
| Benchmark | Claude 3 Haiku | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 36.3% | 90.8% |
| LMArena Expert | 1148 | 1491 |
| SimpleQA Verified | — | 34.9% |
| Confabulations | 34.2% | — |
| Vectara Hallucination Rate | — | 10.8% |
| MMLU | 73.8% | — |
Multimodal Kimi K2.6 leads
Claude 3 Haiku: 23.1 (#123), Kimi K2.6: 31.8 (#95)
| Benchmark | Claude 3 Haiku | Kimi K2.6 |
|---|---|---|
| LMArena Vision | 950 | 1282 |
| Blueprint-Bench 2 | — | 3.9% |
| Furniture Assembly | — | 21.7% |
| LMArena Document | — | 1451 |
| ScienceQA | 72% | — |
Multilingual Kimi K2.6 leads
Claude 3 Haiku: 36.1 (#238), Kimi K2.6: 55.1 (#35)
| Benchmark | Claude 3 Haiku | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | 1178 | 1446 |
| LMArena Chinese | 1155 | 1521 |
| LMArena French | 1194 | 1471 |
| LMArena German | 1174 | 1449 |
| LMArena Japanese | 1102 | 1442 |
| LMArena Korean | 1108 | 1428 |
| LMArena Russian | 1204 | 1446 |
| LMArena Spanish | 1165 | 1463 |
Instruction Following Kimi K2.6 leads
Claude 3 Haiku: 61.5 (#242), Kimi K2.6: 76.5 (#40)
| Benchmark | Claude 3 Haiku | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | 1174 | 1451 |
Long Context Kimi K2.6 leads
Claude 3 Haiku: 36.1 (#232), Kimi K2.6: 44.9 (#50)
| Benchmark | Claude 3 Haiku | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | 1191 | 1468 |
Writing & Preference Kimi K2.6 leads
Claude 3 Haiku: 29.8 (#286), Kimi K2.6: 68.7 (#25)
| Benchmark | Claude 3 Haiku | Kimi K2.6 |
|---|---|---|
| LMArena Text | 1195 | 1455 |
| LMArena Creative Writing | 1157 | 1434 |
| EQ-Bench Creative Writing | 717 | 1725 |
| LMArena Multi-Turn | 1190 | 1453 |
| EQ-Bench 4 | — | 1202 |
Frequently asked questions
Is Claude 3 Haiku better than Kimi K2.6?
Kimi K2.6 is the stronger model overall, scoring 47.5 to 25.7 on the Noometry Index.
Is Claude 3 Haiku or Kimi K2.6 better for coding?
Kimi K2.6 scores higher on coding benchmarks: 50.5 versus 26.1 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and Kimi K2.6 share?
25 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Kimi K2.6 has 51.