What are the hardware requirements for Kimi K2.7 Code?

Kimi K2.7 Code requires approximately 344 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Kimi K2.7 Code and how large is it?

Kimi K2.7 Code is developed by moonshotai, a model with 1058.6b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Kimi K2.7 Code?

Kimi K2.7 Code is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Kimi K2.7 Code by moonshotai — VRAM 344GB

Name: Kimi K2.7 Code
Author: moonshotai

Kimi K2.7 Code is an open-source language model by moonshotai. Features: 1058.6b LLM, VRAM: 344GB, License: other, LLM Explorer Score: 0.47.

Compressed-tensors Conversational Custom code Eval-results Feature-extraction Image-text-to-text Kimi k25 Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/moonshotai/Kimi-K2.7-Code

Kimi K2.7 Code Parameters and Internals

LLM Name	Kimi K2.7 Code
Repository 🤗	https://huggingface.co/moonshotai/Kimi-K2.7-Code
Model Size	1058.6b
Required VRAM	344 GB
Updated	2026-07-30
Maintainer	moonshotai
Model Type	kimi_k25
Model Files	1.0 GB: 1-of-64 9.8 GB: 2-of-64 9.8 GB: 3-of-64 9.8 GB: 4-of-64 9.8 GB: 5-of-64 9.8 GB: 6-of-64 9.8 GB: 7-of-64 9.8 GB: 8-of-64 9.8 GB: 9-of-64 9.8 GB: 10-of-64 9.8 GB: 11-of-64 9.8 GB: 12-of-64 9.8 GB: 13-of-64 9.8 GB: 14-of-64 9.8 GB: 15-of-64 9.8 GB: 16-of-64 9.8 GB: 17-of-64 9.8 GB: 18-of-64 9.8 GB: 19-of-64 9.8 GB: 20-of-64 9.8 GB: 21-of-64 9.8 GB: 22-of-64 9.8 GB: 23-of-64 9.8 GB: 24-of-64 9.8 GB: 25-of-64 9.8 GB: 26-of-64 9.8 GB: 27-of-64 9.8 GB: 28-of-64 9.8 GB: 29-of-64 9.8 GB: 30-of-64 9.8 GB: 31-of-64 9.8 GB: 32-of-64 9.8 GB: 33-of-64 9.8 GB: 34-of-64 9.8 GB: 35-of-64 9.8 GB: 36-of-64
Model Architecture	KimiK25ForConditionalGeneration
License	other

Rank the Kimi K2.7 Code capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Quantized Models of the Kimi K2.7 Code

Model	Likes	Downloads	VRAM
Kimi K2.7 Code GGUF	190	141120	1 GB
Kimi K2.7 Code Mlx DQ3 K M Q8	5	3865	186 GB

Best Alternatives to Kimi K2.7 Code

Best Alternatives	Context / RAM	Downloads	Likes
Kimi K2.6	0K / 334.2 GB	857119	1584
Kimi K2.5	0K / 344 GB	1354033	2822
Kimi K2.6	0K / 334.2 GB	2020	18
Kimi K2.7 Code	0K / 344 GB	101	21
Kimi K2.5	0K / 344 GB	56	14

Note: green Score (e.g. "73.2") means that the model is better than moonshotai/Kimi-K2.7-Code.