No, normal Kimi Code use stores client data, not full model weights
Data locations docs
Why does usage feel larger after switching?
Cache and context behavior can change
Kimi Code model configuration docs
Plain answer
There are two size questions people mix together. Kimi K3 is a 2.8T total-parameter model with a 1M-token context window in official model materials. Kimi Code is the coding tool surface that can call model IDs such as k3, k3-256k, kimi-for-coding, and kimi-for-coding-highspeed.
If you install Kimi Code locally, you are not downloading full Kimi K3 weights. Local storage is for the CLI, config, sessions, credentials, plugins, logs, and history.
AI handoff prompt and permissions
Copy this prompt into Kimi Code, K3Nova, or another AI agent. Keep approval manual for file writes, shell commands, account actions, and secrets.
Copyable AI prompt
Use this to give the agent the task and safety boundary in one message.
You are my AI agent for this task: separate Kimi K3 model size from Kimi Code local storage and tell me what data can grow on disk.
Start by restating the goal and the permissions you need.
Use official docs or the files I provide before making claims.
Give me a direct answer first, then a short table or checklist.
If commands are needed, show exact copyable commands without a shell prompt.
Ask before writing files, running shell commands, deleting or moving data, logging in, spending money, changing account settings, or handling API keys.
Stop and ask me when a step requires secrets, payment, account access, destructive cleanup, or a permission broader than the task.
Recommended AI permissions
Permission
Give AI
Why
Public research
Allow official docs, cited public pages, and web search results.
Needed to refresh facts without touching private data.
Local files
Do not allow local file access unless you name a folder for inspection.
Most research pages do not need your machine.
Write or shell
Ask before Write, Edit, Bash, or any command execution.
Keeps the task reviewable.
Sensitive actions
Deny login, payment, API keys, account changes, and private screenshots.
Those actions need a human decision.
How to read the facts
Kimi K3 size
Official materials list 2.8T total parameters, 104B activated parameters, and a 1M context window.
Kimi Code size
The local tool stores runtime data and session history. The model itself is reached through configured Kimi access.
Quota size
K3 1M context and k3-256k have different consumption implications, so choose the context window that matches the task.
Quick scan cards
How large is Kimi K3?
2.8T total parameters Kimi K3 official repository
How much context can K3 use?
Up to 1M tokens in supported contexts Kimi API and Kimi Code model docs
Does Kimi Code download weights?
No, normal Kimi Code use stores client data, not full model weights Data locations docs
Why does usage feel larger after switching?
Cache and context behavior can change Kimi Code model configuration docs
FAQ
Is Kimi K3 Code a separate 2.8T download?
No. Kimi Code is the tool surface; Kimi K3 is the model it can call when configured.
Why does Kimi Code use local disk?
It stores sessions, config, credentials, plugins, logs, and history.
Should I use k3 or k3-256k?
Use k3 when you need 1M context and have access; use k3-256k when 256k context is enough.
K3Nova