Frequently asked questions

NeonCodex AI is an AI workspace that gives you access to Claude Sonnet 4.6, Claude Opus 4.6, GPT-5.5, Gemini 3.1 Pro, and free models like Qwen3 Coder 480B in one place. You can run tasks, analyze files, write code, do research, and automate workflows — all from a single dashboard or via API.

No. The free plan requires only an email address. You get 10 tasks per day on free models (Qwen3 Coder and Gemma 4 31B) without any payment information.

Create a free account at neoncodex.io/signup, verify your email, and you can immediately start running tasks. The onboarding flow walks you through your first task in under 2 minutes.

Auto routing analyzes your prompt and automatically selects the best model for that task. Coding questions route to Claude Sonnet 4.6 or Qwen3 Coder. Research goes to Gemini 3.1 Pro (2M context). Quick questions go to faster, cheaper models. You can always override this and pick a model manually.

Free plan: Qwen3 Coder 480B, Gemma 4 31B, and Nemotron Super 120B. Pro plan: Claude Haiku 4.5, Claude Sonnet 4.6, Claude Sonnet 5, Claude Opus 4.8, Claude Fable 5, GPT-5.5, Gemini 3.1 Pro (2M context), DeepSeek V4 Pro, GLM-5.2, Kimi K2.7 Code, and all free models.

It depends on the model. Gemini 3.1 Pro has a 2 million token context window — enough for entire codebases. Claude Sonnet 4.6 and GPT-5.5 support 200K tokens. Free models support 32K-131K tokens. File uploads are handled separately and don't count against the context limit in the same way.

Yes. Go to Settings → API Keys and add your own Anthropic, OpenAI, or OpenRouter keys. When set, your tasks use your keys instead of the platform keys — useful if you already have API access or want to control costs directly.

Yes. Upload images (PNG, JPG, WEBP, GIF) and they're sent to vision-capable models. Claude Sonnet 4.6 and GPT-5.5 can analyze images, screenshots, diagrams, and designs. The Image→Code feature specifically converts design screenshots to working code.

Images (all major formats), PDFs, code files, text files, CSV/spreadsheets, JSON, Markdown, and ZIP archives. ZIP files are automatically extracted and the AI reads the directory structure. Maximum 100MB per file, up to 5 files per task on Pro.

Free: 10 tasks/day, free models only, no credit card. Pro: ₹2,499/month ($29 equivalent) — includes Claude, GPT-5.5, Gemini 3.1 Pro, file uploads, memory, and 50 tasks/day. 6-month and annual plans available at 14% and 31% discount respectively.

We accept all major credit cards, debit cards, UPI, net banking, and EMI options via Cashfree. International cards are supported.

We offer a 7-day refund on first-time subscriptions if you're not satisfied. Contact [email protected] within 7 days of your first payment.

Yes. A PDF invoice is automatically emailed after every successful payment. You can also download past invoices from Settings → Billing.

Your account automatically returns to the free plan. Your task history and settings are preserved. You can re-subscribe at any time to restore Pro access.

Create a team workspace, invite members by email, assign roles (owner, admin, member, viewer), and share projects, knowledge bases, and tasks. All team members get Pro access when the team is on a Pro plan.

Owner: full control, billing access. Admin: invite members, manage projects. Member: create tasks, view all team content. Viewer: read-only access to team tasks and projects.

Yes. Team Knowledge Bases let you store SOPs, guidelines, and context that the AI automatically references when team members run tasks. Add documents once, benefit from them in every task.

Yes. The NeonCodex REST API lets you create tasks, stream output, upload files, and manage conversations programmatically. API keys start with nck_live_. Full documentation at neoncodex.io/developer.

Yes. Official JavaScript/TypeScript SDK: npm install @neoncodex/sdk. Official Python SDK: pip install neoncodex. Both support synchronous and async streaming.

Yes: npm install -g neoncodex. Then neoncodex run "your prompt" or pipe stdin directly. Supports all models, file attachments, and output saving.

Yes. Search "NeonCodex Assistant" in the VS Code Extension Marketplace. Features include sidebar chat, right-click actions on selected code (explain, fix, test, security scan), and keyboard shortcuts.

Yes. Your tasks and files are associated only with your account. We don't use your task content to train AI models. See our Privacy Policy for full details.

Task data and files are stored on servers in India. API keys you provide are encrypted at rest using AES-256. We never store third-party API keys in plaintext.

Yes. Go to Settings → Account → Delete Account. This immediately anonymizes all personal data and deletes your task history. The process is permanent and GDPR-compliant.

Yes. Enable two-factor authentication via Settings → Security → Two-factor authentication. We support TOTP (Google Authenticator, Authy, and any standard TOTP app). Backup codes are provided during setup.