Moonshot AIproprietary
Released: 2025-01-25Kimi k1.5
Moonshot AI 2M token context reasoning model trained with reinforcement learning for long-document mathematics and coding.
Intelligence Index
84.0/ 100
Calibrated multi-domain compositeThroughput Speed
48tok/sec
Output streaming throughputTime To First Token
620ms
Initial chunk server latencyLiveBench Contam-Free
70.5%
September 2026 suite scoreSourced Benchmark Evaluations (61)
Verified performance across authoritative benchmarks with provenance tracking.| Benchmark | Status | Score | Trust | Date | Source Type | Provenance |
|---|---|---|---|---|---|---|
| ARC-AGI-3 | Active | 55.3% | 93 | 2025-02-24 | independent | Source ↗ |
| ARC-AGI-2 | Nearing Saturation | 66.3% | 75 | 2025-02-17 | independent | Source ↗ |
| FrontierMath | Active | 24.1% | 98 | 2025-02-14 | independent | Source ↗ |
| PutnamBench | Active | 44.7% | 90 | 2025-02-13 | independent | Source ↗ |
| Omni-MATH | Deprecated | 63.6% | 10 | 2025-02-11 | independent | Source ↗ |
| SWE-Lancer | Deprecated | 48.3% | 20 | 2025-02-11 | independent | Source ↗ |
| Soohak | Active | 65.4% | 98 | 2025-02-09 | independent | Source ↗ |
| FrontierCode | Active | 41.2% | 98 | 2025-02-09 | independent | Source ↗ |
| Humanity's Last Exam | Active | 37.4% | 93 | 2025-02-07 | independent | Source ↗ |
| StructFlowBench | Nearing Saturation | 65.7% | 75 | 2025-02-07 | independent | Source ↗ |
| AIME | Nearing Saturation | 81.7% | 75 | 2025-02-07 | independent | Source ↗ |
| OlympiadBench | Active | 70.5% | 80 | 2025-02-06 | independent | Source ↗ |
| FireBench | Active | 64.9% | 95 | 2025-02-05 | independent | Source ↗ |
| Terminal-Bench | Active | 49.7% | 80 | 2025-02-05 | independent | Source ↗ |
| MathArena | Active | 68.8% | 99 | 2025-02-05 | independent | Source ↗ |
| OpenAI-MRCR | Active | 70.6% | 80 | 2025-02-04 | independent | Source ↗ |
| CodeContests | Deprecated | 46.2% | 10 | 2025-02-04 | independent | Source ↗ |
| SimpleQA | Active | 43.7% | 90 | 2025-02-04 | independent | Source ↗ |
| ComplexBench | Nearing Saturation | 63.4% | 80 | 2025-02-03 | independent | Source ↗ |
| SWE-bench Pro | Active | 44% | 90 | 2025-02-03 | independent | Source ↗ |
| LongBench-Pro | Active | 68.9% | 80 | 2025-02-03 | independent | Source ↗ |
| MultiPL-E | Deprecated | 65.3% | 10 | 2025-02-02 | independent | Source ↗ |
| ZeroSCROLLS | Nearing Saturation | 67.2% | 78 | 2025-02-02 | independent | Source ↗ |
| AgentIF | Nearing Saturation | 52.6% | 80 | 2025-02-01 | independent | Source ↗ |
| MATH | Saturated | 84.3% | 25 | 2025-02-01 | independent | Source ↗ |
| LiveCodeBench | Saturated | 58.2% | 39 | 2025-02-01 | independent | Source ↗ |
| InfiniteBench | Nearing Saturation | 71.4% | 65 | 2025-02-01 | independent | Source ↗ |
| LiveBench (Continuous) | Active | 70.5% | 94 | 2025-02-01 | independent | Source ↗ |
| MMLU-Pro | Nearing Saturation | 77.3% | 75 | 2025-01-31 | independent | Source ↗ |
| Mercury | Deprecated | 62.5% | 10 | 2025-01-31 | independent | Source ↗ |
| SCROLLS | Nearing Saturation | 67.2% | 65 | 2025-01-31 | independent | Source ↗ |
| RULER | Nearing Saturation | 77.3% | 84 | 2025-01-30 | independent | Source ↗ |
| TabMWP | Saturated | 79.1% | 25 | 2025-01-30 | independent | Source ↗ |
| CyberSecEval | Deprecated | 63.9% | 10 | 2025-01-30 | independent | Source ↗ |
| MMLU-Redux | Saturated | 82.3% | 25 | 2025-01-29 | independent | Source ↗ |
| QuALITY | Nearing Saturation | 73.9% | 78 | 2025-01-29 | independent | Source ↗ |
| SWE-bench Verified | Saturated | 55.4% | 25 | 2025-01-29 | independent | Source ↗ |
| MGSM | Nearing Saturation | 81.7% | 65 | 2025-01-28 | independent | Source ↗ |
| LongBench | Nearing Saturation | 72.2% | 65 | 2025-01-28 | independent | Source ↗ |
| SWE-bench | Deprecated | 48.3% | 10 | 2025-01-28 | independent | Source ↗ |
| AGIEval | Saturated | 73.1% | 25 | 2025-01-27 | independent | Source ↗ |
| QASPER | Nearing Saturation | 69.7% | 65 | 2025-01-27 | independent | Source ↗ |
| TruthfulQA | Saturated | 73.9% | 25 | 2025-01-27 | independent | Source ↗ |
| SuperGLUE | Saturated | 86.7% | 25 | 2025-01-26 | independent | Source ↗ |
| GSM-Hard | Deprecated | 82.6% | 10 | 2025-01-26 | independent | Source ↗ |
| MT-Bench | Nearing Saturation | 8% | 75 | 2025-01-25 | independent | Source ↗ |
| BIG-Bench Hard | Saturated | 83.3% | 25 | 2025-01-25 | independent | Source ↗ |
| CodeXGLUE | Deprecated | 66.7% | 23 | 2025-01-25 | independent | Source ↗ |
| MBPP+ | Nearing Saturation | 72.4% | 65 | 2025-01-24 | independent | Source ↗ |
| WinoGrande | Saturated | 88.4% | 28 | 2025-01-24 | independent | Source ↗ |
| IFEval | Deprecated | 75.7% | 25 | 2025-01-24 | independent | Source ↗ |
| GPQA | Nearing Saturation | 73.9% | 78 | 2025-01-24 | independent | Source ↗ |
| HumanEval+ | Saturated | 76.7% | 25 | 2025-01-23 | independent | Source ↗ |
| Needle-in-a-Haystack | Saturated | 96.8% | 44 | 2025-01-23 | independent | Source ↗ |
| CommonsenseQA | Saturated | 89.3% | 28 | 2025-01-23 | independent | Source ↗ |
| GSM8K | Saturated | 92.9% | 25 | 2025-01-22 | independent | Source ↗ |
| GLUE | Saturated | 91.8% | 28 | 2025-01-22 | independent | Source ↗ |
| MBPP | Saturated | 75.3% | 25 | 2025-01-22 | independent | Source ↗ |
| HumanEval | Saturated | 80.9% | 25 | 2025-01-21 | independent | Source ↗ |
| MMLU | Saturated | 85.7% | 25 | 2025-01-21 | independent | Source ↗ |
| ARC-AGI | Saturated | 79.9% | 25 | 2025-01-21 | independent | Source ↗ |
OpenAI Compatible API ExecutionAPI Ready
from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(
model="kimi-k1-5",
messages=[{"role": "user", "content": "Evaluate multi-step logic problem"}],
temperature=0.2
)
print(response.choices[0].message.content)Evaluation & Sourcing Notes
Scores listed for Kimi k1.5 represent verified evaluations extracted from official research papers, independent evaluation suites (HELM, LMSYS, OpenCompass, LiveBench), and verified audit reports.
All benchmarks marked as saturated or deprecated reflect historical performance where the benchmark no longer provides active discriminative power.