Model comparison
Claude 3.5 Haiku vs Kimi K2.6
Kimi K2.6 is the stronger model overall, scoring 47.5 to 28.9 on the Noometry Index.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 1 category and Kimi K2.6 in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.6 leads 56.8 to 13.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.3% for Claude 3.5 Haiku and 96.1% for Kimi K2.6.
- Kimi K2.6 has downloadable open weights; the other is API-only.
Side by side
| Claude 3.5 Haiku | Kimi K2.6 | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 28.9 | 47.5 |
| Released | 2024-10-22 | 2026-04-20 |
| Weights | Proprietary | Open |
| Context window | — | 262K |
| Max output | — | 262K |
| Input $ / M tokens | — | $0.95 |
| Output $ / M tokens | — | $4 |
| Results tracked | 49 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.6 leads
Claude 3.5 Haiku: 32.6 (#258), Kimi K2.6: 50.5 (#41)
| Benchmark | Claude 3.5 Haiku | Kimi K2.6 |
|---|---|---|
| SciCode | 27.4% | 53.5% |
| WeirdML | 30.7% | 55.9% |
| LMArena Coding | 1287 | 1488 |
| SWE-bench Verified | — | 76.7% |
| Aider Polyglot | 28% | — |
| LMArena WebDev | — | 1509 |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 59% | — |
| CadEval | 32% | — |
| ALE-Bench | — | 1,093 |
Agentic & Tool Use Claude 3.5 Haiku leads
Claude 3.5 Haiku: 27.7 (#90), Kimi K2.6: 21.8 (#128)
| Benchmark | Claude 3.5 Haiku | Kimi K2.6 |
|---|---|---|
| OSWorld 2.0 | — | 4.6% |
| BALROG | 19.3% | — |
| ExploitBench | — | 18.4% |
| GBAEval | — | 0.9% |
| GDP.pdf | — | 12% |
| Vending-Bench 2 | — | 6,205 |
Reasoning Kimi K2.6 leads
Claude 3.5 Haiku: 17.1 (#286), Kimi K2.6: 39.6 (#51)
| Benchmark | Claude 3.5 Haiku | Kimi K2.6 |
|---|---|---|
| CritPt | 0% | 8% |
| LMArena Hard Prompts | 1251 | 1470 |
| DTBench | 56.7% | 90.9% |
| Epoch Capabilities Index | 127.15 | 151.03 |
| NYT Connections (extended) | — | 87.2% |
| Chess Puzzles | — | 26% |
| EBR-Bench | — | 2.4% |
| LiveBench Reasoning | 28.1% | — |
| Mystery Game Puzzles | — | 18% |
| LiveBench Data Analysis | 48.5% | — |
| LMCA | — | 37.3% |
| LiveBench | 43.5% | — |
Math Kimi K2.6 leads
Claude 3.5 Haiku: 13.7 (#294), Kimi K2.6: 56.8 (#34)
| Benchmark | Claude 3.5 Haiku | Kimi K2.6 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 96.1% |
| LMArena Math | 1244 | 1475 |
| FrontierMath (Feb 2025 set) | 0.3% | 39% |
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 25.6% |
| MathArena Final-Answer Competitions | — | 72.9% |
| ProofBench | — | 16% |
| Omni-MATH | 22.4% | — |
| LiveBench Math | 35.5% | — |
| MATH Level 5 | 46.4% | — |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge Kimi K2.6 leads
Claude 3.5 Haiku: 18.7 (#276), Kimi K2.6: 53.9 (#48)
| Benchmark | Claude 3.5 Haiku | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 38.1% | 90.8% |
| LMArena Expert | 1208 | 1491 |
| SimpleQA Verified | — | 34.9% |
| MMLU-Pro | 60.5% | — |
| Confabulations | 36.7% | — |
| Vectara Hallucination Rate | — | 10.8% |
| GPQA (HELM) | 36.3% | — |
| MMLU | 74.3% | — |
Multimodal Kimi K2.6 leads
Claude 3.5 Haiku: 26.3 (#113), Kimi K2.6: 31.8 (#95)
| Benchmark | Claude 3.5 Haiku | Kimi K2.6 |
|---|---|---|
| LMArena Vision | 1092 | 1282 |
| GeoBench | 34% | — |
| Blueprint-Bench 2 | — | 3.9% |
| Furniture Assembly | — | 21.7% |
| LMArena Document | — | 1451 |
Multilingual Kimi K2.6 leads
Claude 3.5 Haiku: 40.2 (#213), Kimi K2.6: 55.1 (#35)
| Benchmark | Claude 3.5 Haiku | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | 1238 | 1446 |
| LMArena Chinese | 1229 | 1521 |
| LMArena French | 1262 | 1471 |
| LMArena German | 1237 | 1449 |
| LMArena Japanese | 1175 | 1442 |
| LMArena Korean | 1172 | 1428 |
| LMArena Russian | 1253 | 1446 |
| LMArena Spanish | 1261 | 1463 |
Instruction Following Kimi K2.6 leads
Claude 3.5 Haiku: 63.1 (#229), Kimi K2.6: 76.5 (#40)
| Benchmark | Claude 3.5 Haiku | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | 1241 | 1451 |
| LiveBench Instruction Following | 61.9% | — |
| IFEval | 79.2% | — |
Long Context Kimi K2.6 leads
Claude 3.5 Haiku: 38.3 (#195), Kimi K2.6: 44.9 (#50)
| Benchmark | Claude 3.5 Haiku | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | 1261 | 1468 |
Writing & Preference Kimi K2.6 leads
Claude 3.5 Haiku: 42.9 (#228), Kimi K2.6: 68.7 (#25)
| Benchmark | Claude 3.5 Haiku | Kimi K2.6 |
|---|---|---|
| LMArena Text | 1255 | 1455 |
| LMArena Creative Writing | 1233 | 1434 |
| EQ-Bench Creative Writing | 1146 | 1725 |
| LMArena Multi-Turn | 1265 | 1453 |
| Short-Story Creative Writing | 73.5% | — |
| WildBench | 76% | — |
| EQ-Bench 4 | — | 1202 |
| LiveBench Language | 35.4% | — |
Frequently asked questions
Is Claude 3.5 Haiku better than Kimi K2.6?
Kimi K2.6 is the stronger model overall, scoring 47.5 to 28.9 on the Noometry Index.
Is Claude 3.5 Haiku or Kimi K2.6 better for coding?
Kimi K2.6 scores higher on coding benchmarks: 50.5 versus 32.6 in the Noometry coding category.
How many benchmarks do Claude 3.5 Haiku and Kimi K2.6 share?
27 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and Kimi K2.6 has 51.