Model comparison
Claude 3 Haiku vs Kimi K2 (Jul 2025)
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 25.9 on the Noometry Index.
Last verified . 23 shared benchmarks.
Summary
- They share 23 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Kimi K2 (Jul 2025) in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2 (Jul 2025) leads 42.7 to 9.8.
- The biggest single-benchmark swing is WeirdML: 9.8% for Claude 3 Haiku and 42.8% for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Haiku | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 25.9 | 41.2 |
| Released | 2024-03-07 | 2025-07-12 |
| Weights | Proprietary | Open |
| Context window | — | 262K |
| Max output | — | 262K |
| Input $ / M tokens | — | $0.57 |
| Output $ / M tokens | — | $2.30 |
| Results tracked | 37 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
Claude 3 Haiku: 26.4 (#325), Kimi K2 (Jul 2025): 42.4 (#102)
| Benchmark | Claude 3 Haiku | Kimi K2 (Jul 2025) |
|---|---|---|
| WeirdML | 9.8% | 42.8% |
| LMArena Coding | 1199 | 1399 |
| SWE-bench Verified (bash only) | — | 63.4% |
| Aider Polyglot | — | 59.1% |
| GSO | — | 4.9% |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| ALE-Bench | — | 597.5 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |
Agentic & Tool Use Not comparable
Claude 3 Haiku: —, Kimi K2 (Jul 2025): 32.4 (#64)
| Benchmark | Claude 3 Haiku | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | — | 35.7% |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| METR Time Horizons | — | 59.2% |
Reasoning Kimi K2 (Jul 2025) leads
Claude 3 Haiku: 16.3 (#307), Kimi K2 (Jul 2025): 23.3 (#179)
| Benchmark | Claude 3 Haiku | Kimi K2 (Jul 2025) |
|---|---|---|
| Kagi LLM Benchmark | 34.2% | 64.4% |
| LMArena Hard Prompts | 1174 | 1384 |
| Epoch Capabilities Index | 118.35 | 146.01 |
| ForecastBench | 53.2 | 60.2 |
| SimpleBench | — | 26.3% |
| DTBench | 50.1% | — |
| LMCA | 8.8% | — |
| WinoGrande | 74.2% | — |
Math Kimi K2 (Jul 2025) leads
Claude 3 Haiku: 9.8 (#319), Kimi K2 (Jul 2025): 42.7 (#83)
| Benchmark | Claude 3 Haiku | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Math | 1188 | 1397 |
| OTIS Mock AIME 2024-2025 | 1.8% | — |
| Omni-MATH | — | 65.4% |
| MATH Level 5 | 14.9% | — |
| FrontierMath (Feb 2025 set) | — | 21.4% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Kimi K2 (Jul 2025) leads
Claude 3 Haiku: 17.3 (#285), Kimi K2 (Jul 2025): 37.3 (#157)
| Benchmark | Claude 3 Haiku | Kimi K2 (Jul 2025) |
|---|---|---|
| Confabulations | 34.2% | 20.4% |
| LMArena Expert | 1148 | 1365 |
| GPQA Diamond | 36.3% | — |
| MMLU-Pro | — | 81.9% |
| Vectara Hallucination Rate | — | 17.9% |
| GPQA (HELM) | — | 65.3% |
| MMLU | 73.8% | — |
Multimodal Not comparable
Claude 3 Haiku: 23.6 (#128), Kimi K2 (Jul 2025): —
| Benchmark | Claude 3 Haiku | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Vision | 950 | — |
| ScienceQA | 72% | — |
Multilingual Kimi K2 (Jul 2025) leads
Claude 3 Haiku: 36.0 (#243), Kimi K2 (Jul 2025): 49.6 (#130)
| Benchmark | Claude 3 Haiku | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1178 | 1372 |
| LMArena Chinese | 1155 | 1415 |
| LMArena French | 1195 | 1379 |
| LMArena German | 1174 | 1387 |
| LMArena Japanese | 1102 | 1349 |
| LMArena Korean | 1109 | 1325 |
| LMArena Russian | 1204 | 1385 |
| LMArena Spanish | 1166 | 1386 |
Instruction Following Kimi K2 (Jul 2025) leads
Claude 3 Haiku: 61.3 (#247), Kimi K2 (Jul 2025): 71.1 (#156)
| Benchmark | Claude 3 Haiku | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Instruction Following | 1173 | 1348 |
| IFEval | — | 85% |
Long Context Kimi K2 (Jul 2025) leads
Claude 3 Haiku: 36.1 (#237), Kimi K2 (Jul 2025): 41.2 (#145)
| Benchmark | Claude 3 Haiku | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Longer Query | 1190 | 1353 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 17.6% |
Writing & Preference Kimi K2 (Jul 2025) leads
Claude 3 Haiku: 29.7 (#291), Kimi K2 (Jul 2025): 62.3 (#78)
| Benchmark | Claude 3 Haiku | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1195 | 1380 |
| LMArena Creative Writing | 1157 | 1350 |
| EQ-Bench Creative Writing | 717 | 1666 |
| LMArena Multi-Turn | 1190 | 1371 |
| Short-Story Creative Writing | — | 85.6% |
| WildBench | — | 86.2% |
Frequently asked questions
Is Claude 3 Haiku better than Kimi K2 (Jul 2025)?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 25.9 on the Noometry Index.
Is Claude 3 Haiku or Kimi K2 (Jul 2025) better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 42.4 versus 26.4 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and Kimi K2 (Jul 2025) share?
23 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Kimi K2 (Jul 2025) has 42.