Yi: yi-lightning API

01.AI's flagship Mixture-of-Experts (MoE) model optimized for low-latency and high-performance. It delivers frontier-level reasoning, math, and coding capabilities through advanced expert segmentation and KV-cache techniques. It is highly competitive in multilingual proficiency and instruction-following, consistently ranking among the top Chinese models on the Chatbot Arena.

Frequently Asked Questions

What is the context window of yi-lightning?

yi-lightning supports a context window of up to 163,840 tokens.

Does yi-lightning support function calling?

Yes. yi-lightning supports tool / function calling.

Does yi-lightning support reasoning?

Yes. yi-lightning is a reasoning-capable model.

What is the knowledge cutoff of yi-lightning?

The knowledge cutoff of yi-lightning is 2025-05-31.