Moonshot AI
Reasoning

Kimi k1.5

Reasoning

Moonshot's reasoning Kimi. RL-trained chain-of-thought competitive with o1 on math benchmarks; long-context strength is the Kimi signature.

Kimi k1.5 is a reasoning AI model from Moonshot AI. It costs $2.50 per million input tokens and $10.00 per million output tokens (blended $4.75/M), with a 128,000-token context window.

RatesUSD per million tokens. Blend is 70% input, 30% output, our calculation.

INPUT
$2.50/M
per million input tokens
OUTPUT
$10.00/M
per million output tokens
BLENDED 70/30
$4.75/M
unchanged since 3 May
CONTEXT
128,000
tokens

Hand-maintained from Moonshot AI's published pricing; no check date on record.

What it is good at
  • Reasoning + long context
  • Strong math
  • Visible thinking
  • Aggressive Chinese pricing
Typical use cases
  • Reasoning agents
  • Long-document STEM analysis

Benchmarks

vs. best public score
MMLU87.4%
Multitask academic knowledge across 57 subjects. Source: Moonshot AI (self-reported), MMLU (EM), read 2026-11-10.
Hand-curated from each provider's published reports and public leaderboards. Methodology varies across sources, treat as directional rather than authoritative.

Also reported by the provider

Results on benchmarks outside the set we compare across models. They are listed for completeness and are deliberately excluded from the bars above and from every ranking on this site, because a score is only comparable against the same benchmark variant. Why benchmark scores do not compare →

  • MATH-500 (EM)96.2%self-reported by Moonshot AI · 2025-01-22 · Long-CoT model. A 500-problem subset, not comparable with the full MATH benchmark.
  • HumanEval-Mul (Pass@1)81.5%self-reported by Moonshot AI · 2025-01 · Kimi k1.5 paper, Table 3 (short-CoT model). HumanEval-Mul is a translation into eight languages (Python, Java, C++, C#, JavaScript, TypeScript, PHP, Bash), not Python-only HumanEval. Not comparable with the HumanEval column, so shown here and never ranked.

More from Moonshot AI

See all 5 →

Frequently asked questions

How much does Kimi k1.5 cost?

Kimi k1.5 costs $2.50 per million input tokens and $10.00 per million output tokens, for a blended reference rate of $4.75 per million tokens.

What is Kimi k1.5's context window?

Kimi k1.5 supports up to 128,000 tokens of context in a single request.

What is Kimi k1.5 best for?

Kimi k1.5 is well suited to Reasoning + long context, Strong math and Visible thinking.

Who makes Kimi k1.5?

Kimi k1.5 is developed and served by Moonshot AI. It was released in Jan 2025.

Compared with

Terms used on this page