Kimi-K2-Instruct
About Kimi-K2-Instruct
Kimi K2 is a Mixture-of-Experts (MoE) foundation model with exceptional coding and agent capabilities, featuring 1 trillion total parameters and 32 billion activated parameters. In benchmark evaluations covering general knowledge reasoning, programming, mathematics, and agent-related tasks, the K2 model outperforms other leading open-source models
Explore how Kimi-K2-Instruct's exceptional coding, agentic capabilities, and deep reasoning can be applied to solve complex, real-world problems.
Software Engineering Agent
Automate complex software tasks from design to deployment, leveraging Kimi-K2's agentic and coding prowess.
Use Case Example:
"An agent autonomously refactored a legacy Java microservice, integrating new API endpoints and generating comprehensive unit tests, reducing manual effort by 70%."
Polyglot Code Analysis
Analyze vast, multi-language codebases to identify vulnerabilities, optimize performance, and ensure architectural consistency.
Use Case Example:
"Scanned an enterprise system (Python, Go, JS) for cross-language dependency issues, pinpointing security patches and preventing potential breaches."
Autonomous Data Pipelines
Design, implement, and monitor complex data ingestion and transformation pipelines, adapting to schema changes and optimizing resource usage.
Use Case Example:
"Automatically generated and deployed a Spark-based ETL pipeline for real-time sensor data, handling schema evolution and optimizing query performance."
Scientific Modeling & Sim
Assist researchers in building, validating, and iterating on complex scientific models, from physics to biological systems.
Use Case Example:
"Collaborated with a materials scientist to simulate molecular dynamics in Rust, iteratively refining parameters and visualizing results, accelerating experimental design."
Intelligent System Audits
Perform deep logical audits of complex systems, including cloud configurations, smart contracts, and regulatory documents.
Use Case Example:
"Audited a Kubernetes cluster configuration against best practices, identifying misconfigurations and generating remediation scripts in YAML."
Metadata
Specification
State
Deprecated
Architecture
Mixture-of-Experts
Calibrated
No
Mixture of Experts
Yes
Total Parameters
1000B
Activated Parameters
32B
Reasoning
No
Precision
FP8
Context length
131K
Max Tokens
131K
Compare with Other Models
See how this model stacks up against others.
Moonshot AI
chat
Kimi-K2.6
Release on: Apr 21, 2026
Total Context:
262K
Max output:
262K
Input:
$
0.77
/ M Tokens
Output:
$
4.0
/ M Tokens
Moonshot AI
chat
Kimi-K2.5
Release on: Jan 30, 2026
Total Context:
262K
Max output:
262K
Input:
$
0.45
/ M Tokens
Output:
$
2.25
/ M Tokens
Moonshot AI
chat
Kimi-K2-Thinking
Release on: Nov 7, 2025
Total Context:
262K
Max output:
262K
Input:
$
0.55
/ M Tokens
Output:
$
2.5
/ M Tokens
Moonshot AI
chat
Kimi-K2-Instruct-0905
Release on: Sep 8, 2025
Total Context:
262K
Max output:
262K
Input:
$
0.4
/ M Tokens
Output:
$
2.0
/ M Tokens
Moonshot AI
chat
Kimi-K2-Instruct
Release on: Jul 13, 2025
Total Context:
131K
Max output:
131K
Input:
$
0.58
/ M Tokens
Output:
$
2.29
/ M Tokens
Moonshot AI
chat
Kimi-Dev-72B
Release on: Jun 19, 2025
Total Context:
131K
Max output:
131K
Input:
$
0.29
/ M Tokens
Output:
$
1.15
/ M Tokens
