#33
Kimi K2.5
kimi-k2.5
0.0
Helpfulness
Instruction Following
Comprehension
Empathy
Creative Writing
Helpfulness
0.0
Instruction Following
0.0
Comprehension
0.0
Empathy
0.0
Creative Writing
0.0
Speed
Avg 65 tok/s
Release Date
January 27, 2026
Lab
Moonshot AI
Type
Open Source
Context Size
256K
Max Output Tokens
4.1K
Cost per 1 million tokens
$0.55 / $2.75
Model Inputs*
Text, Images, Video
Model Outputs*
Text
Tool Calling*
Enabled
Overall Assistant Score
An average score combining the 5 main categories.
85.92 pts
Rank #33
54th Percentile
0
Novice
33
Capable
66
Proficient
100
Expert
Kimi K2.5 is a highly efficient, open-source multimodal model from Moonshot AI, utilizing a 1-trillion parameter MoE architecture. It excels in instruction following (87.5) and comprehension (88.2), making it a top-tier choice for complex reasoning and agentic workflows. While its empathy scores (72.5) are more modest, its primary strengths lie in its cost-effectiveness and 'Agent Swarm' technology, which enables superior performance in parallel task coordination compared to proprietary models. Its native support for video and visual coding makes it a versatile tool for developers and enterprises seeking high performance at a lower price point.

Overall Score vs Price

Price axis is logarithmic (USD per 1M input tokens) · Higher score and lower price is better

Ringed point is this model

Anthropic
DeepSeek
Google
Meta
Moonshot
OpenAI
SpaceX AI
Zhipu AI

Intelligence

Overall Score · Higher is better

Speed

Output tokens per second · Higher is better

Price

USD per 1M input tokens · Lower is better

Moonshot Models

Overall Score · Same provider

Moonshot Family

Overall Score · Same model family

Closest Rivals

Overall Score · Nearest by overall rank