Moonshot AIproprietary
Released: 2025-01-20

Kimi 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 composite
Throughput Speed
48tok/sec
Output streaming throughput
Time To First Token
620ms
Initial chunk server latency
LiveBench Contam-Free
70.5%
September 2026 suite score

Sourced Benchmark Evaluations (2)

Verified performance across authoritative benchmarks with provenance tracking.
BenchmarkStatusScoreTrustDateSource TypeProvenance
ARC-AGI-2Nearing Saturation66.3%
75
2025-02-17independentSource ↗
SWE-bench VerifiedSaturated54%
25
2025-01-29independentSource ↗
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.