Powered exclusively by NVIDIA NIM

Your AI pair programmer, inside VS Code.

Chat, explain, refactor, and ship code with state-of-the-art models — Qwen, GLM, Nemotron, DeepSeek and more — all through a single NVIDIA API key. Plus an autonomous agent that reads, writes, and runs your workspace.

Install for VS Code Quick start →
Free tier available Keys in OS keychain No telemetry
N NIM Code ☰ ＋
YOU
Add input validation to the signup handler and write a test for it.
NIM CODE · Qwen3 Next 80B
I'll read the handler, add a Zod schema, then generate tests.
🔧 read_file(src/api/signup.ts)
🔧 write_file(src/api/signup.ts)
🔧 write_file(src/api/signup.test.ts)
Done — added validation and 6 test cases
💬 Chat⚡ Agent
Ask NIM Code… (/ for commands) Send
Models on tap qwen3-next-80b kimi-k2.6 nemotron-3-super-120b glm-5.1 deepseek-v4-flash + any NIM model
12
Agent tools
13
Slash commands
131K
Token context
∞
Custom models
Everything you need

Built for real coding work

A focused, fast assistant that lives in your sidebar and tracks your editor theme — no context switching, no copy-paste.

Multi-model chat

Switch between curated NVIDIA NIM models from a dropdown, or add your own model ID with the + button — no extension update needed.

Autonomous agent mode

Hand off multi-step tasks. The agent reads, searches, writes and refactors files, runs terminal commands, queries git, and opens files in your editor — showing each tool call as it works. A diff preview gates every file write so you stay in control.

Slash commands

Type / for context-aware actions on your selection — explain, fix, review, refactor, test, document, and more.

Token streaming

Responses stream token-by-token in both chat and agent mode, with full cancel support mid-generation.

Secure key storage

Your NVIDIA NIM API key lives in the OS keychain via VS Code SecretStorage — never in settings files or on disk.

Persistent sessions

Every conversation is saved locally. Rename, delete, and switch between sessions from the history list anytime.

Agent mode

12 built-in tools

The agent loop calls these one at a time, reasoning between each step — so it stays grounded in your actual codebase.

Files
read_file
Read any file in your workspace
Files
list_files
List files and folders in a directory
Files
write_file
Create or overwrite a file (with diff preview)
Files
delete_file
Delete a file from the workspace
Files
rename_file
Rename or move a file within the workspace
Search
search_codebase
Ripgrep-powered search with line numbers
Terminal
run_terminal
Execute any shell command (30s timeout)
Git
git_status
Show modified, staged & untracked files
Git
git_diff
Show uncommitted changes, scoped or staged
Git
git_log
Show recent commit history
Editor
get_diagnostics
Read errors & warnings from the Problems panel
Editor
open_file_in_editor
Open a file and jump to a line
Slash commands

One keystroke to context

Highlight code, type a slash command, and the right system prompt is sent automatically with your selection attached.

/explainExplain the code /fixFind & fix bugs /reviewQuality review /refactorRestructure /optimizePerformance /testGenerate tests /documentAdd docs /debugDiagnose errors /commitCommit message /summarySummarize a file /securityOWASP review /migrateVersion migration /prPR title & body
Supported models

Best-in-class, your choice

Curated coding models ship in the picker. Add any model served by NVIDIA NIM — or a self-hosted deployment — in seconds.

DEFAULT
Qwen3 Next 80B
qwen/qwen3-next-80b-a3b-instruct
Fast MoE model for coding and reasoning.
131K context
Kimi K2.6
moonshotai/kimi-k2.6
Long-context multilingual model.
131K context
Nemotron Super 120B
nvidia/nemotron-3-super-120b-a12b
NVIDIA-tuned reasoning and code.
131K context
GLM 5.1
z-ai/glm-5.1
ZhipuAI — strong reasoning and code.
131K context
DeepSeek V4 Flash
deepseek-ai/deepseek-v4-flash
Fast, efficient coding assistant.
131K context

Add more from build.nvidia.com via the + button, or point nimcode.baseUrl at a self-hosted NIM endpoint.

Quick start

Running in four steps

From zero to your first completion in under two minutes.

1

Get an API key

Sign up at build.nvidia.com → API Keys → Generate Personal Key. Copy your nvapi-… key.

2

Install NIM Code

Search "NIM Code" in the Extensions panel, or grab it from the VS Code Marketplace.

3

Set your key

Run NIM Code: Set NVIDIA NIM API Key from the Command Palette and paste your key.

4

Start coding

Click the N icon in the Activity Bar or press Ctrl+Shift+N. Done.

Configurable & private

Tune it. Trust it.

Sensible defaults out of the box, with full control over models, timeouts, and where your data goes.

settings.json
{
  // NVIDIA NIM endpoint — change for self-hosted
  "nimcode.baseUrl": "https://integrate.api.nvidia.com/v1",

  // Default model for new sessions
  "nimcode.defaultModel": "qwen/qwen3-next-80b-a3b-instruct",

  // Per-request timeout (ms)
  "nimcode.requestTimeoutMs": 120000,

  // Require diff approval before writes
  "nimcode.agentAutoApplyWrites": false,

  // Bring your own models
  "nimcode.customModels": [
    { "id": "meta/llama-3.3-70b-instruct" }
  ]
}
API key
Stored in the OS keychain. Never written to disk or settings.
No telemetry
Chat data goes only to NVIDIA NIM (or your custom baseUrl).
Strict CSP webview
The UI makes zero external network requests of its own.
Workspace-scoped
All agent file operations stay inside your project root.
On-premise ready
Point baseUrl at localhost — everything works identically.

Ship faster with NIM Code

Free NVIDIA NIM tier, secure by default, and it tracks your editor theme. Install and run your first prompt in minutes.

Install for VS Code Contact us