Discover Kimi Linear 48B A3B Instruct by MoonshotAI

Kimi Linear excels in handling long-context tasks with efficiency.

Input: text Output: text Context: 1,048,576 tokens Release: 2025-11-08

Introducing the Kimi Linear 48B A3B Instruct model by MoonshotAI, released on November 8, 2025. This model features a remarkable context length of up to 1 million tokens, making it ideal for both short and long text tasks. With its innovative Kimi Delta Attention (KDA) mechanism, Kimi Linear optimizes memory usage and enhances decoding throughput significantly. It supports text input and output modalities, ensuring versatility in applications. The model’s design allows for efficient processing in reinforcement learning and other complex scenarios, reducing the need for extensive key-value caches while maintaining high performance.

Use Cases

Here are a few ways teams apply MoonshotAI: Kimi Linear 48B A3B Instruct in practice—from fast drafting to multimodal understanding. Adapt these ideas to your workflow.

Efficiently process long documents

Enhance performance in RL applications

Streamline text generation tasks

Support complex dialogue systems

Key Features

A quick look at the capabilities that make this model useful in real projects.

Hybrid linear attention architecture

Handles up to 1M tokens efficiently

Reduces KV cache needs by 75%

Boosts decoding throughput by 6x

Optimized for reinforcement learning tasks

Specs

Overview

Vendor
moonshotai

Model ID
moonshotai/kimi-linear-48b-a3b-instruct

Release
2025-11-08

Modalities & context

Input
text

Output
text

Context
1,048,576 tokens

Parameters & defaults

Supported parameters: frequency_penalty, logit_bias, logprobs, max_tokens, min_p, presence_penalty, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, top_k, top_logprobs, top_p

Defaults: temperature 0.2, top_p 0.95

Benchmark tests: MoonshotAI: Kimi Linear 48B A3B Instruct

We ran this model against a few representative prompts to show its range. Review the outputs below and be the judge.

Text

Prompt:

Write 150 words on how AI might positively upend work, leisure and creativity

AI could revolutionize work, leisure, and creativity by automating routine tasks, freeing humans to focus on higher-value, more meaningful activities. In work, AI can enhance productivity through data analysis, predictive modeling, and personalized assistance, allowing professionals to make better decisions faster. This shift could lead to shorter workweeks and more flexible schedules, improving work-life balance. In leisure, AI can tailor experiences—whether in gaming, travel, or entertainment—to individual preferences, making downtime more enjoyable and personalized. Creatively, AI tools can assist artists, writers, and designers by generating ideas, offering feedback, or even co-creating content, expanding the boundaries of human imagination. Rather than replacing human ingenuity, AI can augment it, fostering collaboration between people and machines. Ultimately, AI has the potential to redistribute time and energy toward innovation, exploration, and deeper human connection—redefining what it means to work, relax, and create.

Run this prompt on Upend.AI

Ready to try MoonshotAI: Kimi Linear 48B A3B Instruct?

Chat with Kimi Linear 48B A3B Instruct

up.end

/ˌəpˈend/

verb

To “upend” means to completely disrupt, overturn, or drastically change the established order or structure of something. It implies a significant shift or alteration that can potentially have far-reaching consequences. When something is upended, it is turned upside down or transformed in a way that challenges conventional norms or expectations. The term often carries a sense of innovation, transformation, and sometimes even a hint of upheaval, indicating that the changes are not just minor adjustments but rather a fundamental reimagining of the status quo.

Privacy Terms