All AI models · LLM benchmarks · Methodology

Moonshot AI

Kimi K2.5

Earlier Kimi frontier model for long-context agents, coding, and multimodal work

Model facts

Context window
262K tokens
Maximum output
262K tokens
Cheapest paid input
$0.3 per 1M tokens
Cheapest paid output
$1.9 per 1M tokens
Open weights
yes
Providers
51

Capabilities and modalities

reasoning, tool calling, structured output, attachments, temperature control, open weights, text, image, video.

Local hardware estimate

At 4K context: Q4_K_M 629.5 GiB, Q8_0 1107.7 GiB, F16 2004.2 GiB working memory.

Parameters
1026.9 billion
Architecture
kimik25
Model license
other

Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.

Artificial Analysis benchmarks

Compare this model on the full LLM benchmark leaderboard.

ConfigurationAAICodingMathOutput tokens/s
Reasoning36.046.8not measurednot measured
Non-reasoning30.1not measurednot measurednot measured

Observed price history

  • 2026-08-24: $0.3 input / $1.9 output per 1M tokens

API providers