Kimi K2 Instruct is a 1 trillion parameter Mixture-of-Experts (MoE) LLM designed for agentic intelligence and advanced tool use.

It excels in handling complex reasoning, coding, tool or function calling, and agent-based tasks, making it particularly strong for workflows that require function/tool integration and autonomous problem solving.

Some other noteworthy use cases of Kimi K2 Instruct include deep document comprehension enabled by its long 128K token context window, and facilitating command-line interactions or automation through native tool usage.

Metric	Value
Parameter Count	1 trillion
Mixture of Experts	Yes
Active Parameter Count	32 billion
Context Length	128,000 tokens
Multilingual	Yes
Quantized*	Yes
Precision*	FP8

*Quantization is specific to the inference provider and the model may be offered with different quantization levels by other providers.