# Kimi Linear 48B A3B Instruct: AI model fact sheet

- **Provider:** moonshotai
- **Released:** 2025-11-10
- **Context window:** 1,048,576 tokens
- **API pricing:** $0.30 / 1M input, $0.60 / 1M output
- **OpenRouter ID:** moonshotai/kimi-linear-48b-a3b-instruct
- **Capabilities:** conversation, reasoning, code-generation, analysis

Kimi Linear is a hybrid linear attention architecture that outperforms traditional full attention methods. Features Kimi Delta Attention (KDA) for efficient memory usage, reducing KV caches by up to 75% and boosting throughput by up to 6x for contexts as long as 1M tokens.

Source: real side-by-side outputs, pricing and specs at https://rival.tips/models/kimi-linear-48b-a3b-instruct