---
title: "Moonshot AI's Kimi K2.7 Code Burns 30% Fewer Tokens on Agentic Tasks"
subtitle: "Moonshot AI's Kimi K2.7 Code brings a 1T-parameter open-weight coding agent with 30% fewer reasoning tokens and forced multi-turn thinking, targeting long-horizon agentic workflows."
url: "https://alphasignal.ai/news/moonshot-ai-s-kimi-k2-7-code-burns-30-fewer-tokens-on-agentic-tasks"
author: "AlphaSignal Newsroom"
author_url: "https://alphasignal.ai/editorial-team"
company: "Unsloth AI"
company_url: "https://alphasignal.ai/company/unslothai"
published: "2026-06-12T18:03:48.000Z"
updated: "2026-06-17T21:33:06.853Z"
type: "Model"
topics:
  - "Development"
  - "Llms"
subtopics:
  - "Code Agents"
  - "Mixture Of Experts"
  - "Long Context"
upvotes: 10
source_url: "https://huggingface.co/unsloth/Kimi-K2.7-Code-GGUF"
comments: 0
truncated: true
---

# Moonshot AI's Kimi K2.7 Code Burns 30% Fewer Tokens on Agentic Tasks

> Moonshot AI's Kimi K2.7 Code brings a 1T-parameter open-weight coding agent with 30% fewer reasoning tokens and forced multi-turn thinking, targeting long-horizon agentic workflows.

Moonshot AI has shipped [Kimi K2.7 Code](https://huggingface.co/moonshotai/Kimi-K2.7-Code), a coding-focused open-weight model built directly on top of K2.6\. It is the fifth major release in the K2 series in under a year, and it arrives with a clear thesis: the bottleneck in agentic coding is not raw capability, it is how many tokens the model burns getting there.

## Not a new base model, a sharper one

This is not a new base model. It is a focused coding upgrade built on the Kimi K2.6 foundation, targeting two things that matter most in production agentic workflows: better output on long-horizon coding tasks and fewer tokens burned getting there. K2.7 Code is purpose-built for coding tasks. For general-purpose work such as writing, analysis, and conversation, Moonshot recommends K2.6, which offers more well-rounded capabilities. 

Moonshot AI shipped Kimi K2.7-Code on June 12, 2026 , a coding-focused successor to Kimi K2.6 that keeps the same 1T-parameter MoE architecture but retrains the reward model and data pipeline around real-world long-horizon software tasks. Moonshot AI, the Beijing-based lab, has been one of the most consistent open-weight shippers in the last year.

_Preview only. Full text requires an AlphaSignal Pro account._