Kimi K2.7-Code

Name: Kimi K2.7-Code
Author: Moonshot AI

Released

2026-06-12

Last refreshed

2026-06-25

Status

Researched 124d ago

DeprecatedOpen sourceCommercial use: permittedMultimodalCodingRAGAgentsLong contextVisionJSON / Tool use

Kimi K2.7-Code is a legacy integration reference; keep it only while you identify a current replacement.

Moonshot AI releases · 8 in the last 12 months · this family litChangelog →

Specifications

Family: Kimi K2
Released: 2026-06-12
Lifecycle: Scheduled retirement: December 31, 2026
Context: 256K
Max output: 64K
Parameters: 1T
Architecture: Mixture of Experts
Specialization: code
Openness: Open source
License: MITOSI-approvedPermitted
Weights: Available
Code: Unknown

Created by

Moonshot AI

Lossless long-context AI innovation

Beijing, China

Founded 2023

Website

Pricing

Output / 1M

$4.00

Input / 1M

$0.950

Cheapest of 1 route · OpenRouter · cache read $0.190

Providers(1)

OpenRouter

View 1 provider route

Links

Website HuggingFace

About

Kimi K2.7-Code is Moonshot AI's coding-focused multimodal model released June 12, 2026, built on Kimi K2.6. Uses the same 1-trillion-parameter MoE architecture (32B active parameters, 384 experts with 8 selected per token, 61 layers) with a 262K context window and MoonViT vision encoder (400M parameters). Reports +21.8% on Moonshot's Kimi Code Bench v2, +11.0% on Program Bench, +31.5% on MLS Bench Lite versus K2.6, with approximately 30% fewer reasoning tokens. Forces thinking mode on by default and preserves reasoning content across multi-turn interactions for agentic use. Available via Kimi platform API and HuggingFace under Modified MIT license.

Providers(1)

Compare API pricing across 1 tracked provider route for input and output tokens, batch, and cached reads when available.

Provider	Input / 1M	Output / 1M	Cache	Route
OpenRouter	$0.950	$4.00	read $0.190	Serverless

Deprecated provider routes(1)

Provider	Status	Catalog ID
Moonshot AI Kimi	DeprecatedScheduled retirement: December 31, 2026	kimi-k2.7-code

Capabilities

VisionMultimodalReasoningJSON / Tool useStructured OutputsPrompt Caching

Benchmark peer barsfor Coding

No task-mapped benchmark peers are available for this model yet.

Benchmark scores(8)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.

Benchmark	Score	Version	Evaluation	Source
Kimi Code Bench v2	62.0	official, Moonshot-reportedObserved 2026-06-12	—	Source
Program Bench	53.6	official, Moonshot-reportedObserved 2026-06-12	—	Source
MLS Bench Lite	35.1	official, Moonshot-reportedObserved 2026-06-12	—	Source
Kimi Claw 24/7 Bench	46.9	official, Moonshot-reportedObserved 2026-06-12	—	Source
MCP-Atlas	76.0	official, Moonshot-reportedObserved 2026-06-12	—	Source
MCP Mark Verified	81.1	official, Moonshot-reportedObserved 2026-06-12	—	Source
GeneBench-Pro	2.3	reasoning_enabledObserved 2026-06-30	—	Source
CursorBench	49.7	CursorBench 3.2Observed 2026-07-18	Configuration: Kimi K2.7 Code Harness: CursorBench 3.2 productized Cursor-agent workflow Evaluator: Cursor Confidence: confirmed Cost/task: $1.43 Tokens/task: 31,247 Steps/task: 58 Notes: Cursor vendor-reported result; not independently reproducible. Results are subject to variance, and small score differences may not be statistically meaningful.	Source