nascoder

Help & FAQ

Last updated May 9, 2026

Welcome to nascoder. This page collects answers to the questions we hear most often. If you cannot find what you need below, email support@nascoder.com and we will respond within one business day.

Getting started

To create an account, visit app.nascoder.com and click Sign up. Authentication is provided by Auth0; you can sign up with an email address and password or with a supported social identity provider. Once signed in, click New chat to open a conversation, type your message, and press Enter to send it. The model selector at the top of the chat lets you choose which model handles the conversation; you can change models mid-conversation, and the new model will see the existing message history.

New accounts start on the Free plan. You can upgrade at any time from Billing.

Models

nascoder gives you access to a curated selection of frontier models, including Claude Sonnet 4.6, GPT-5, and Llama, served through DigitalOcean Gradient AI’s serverless inference API. The models available to you depend on your subscription tier; the model selector visually indicates which models are available on your plan and which require an upgrade.

If you are unsure which model to choose, select Auto. Auto routing picks a suitable model based on length, complexity, and the kind of task; for short questions it favors a fast, low-cost model, and for longer or more demanding requests it routes to a stronger model.

  • Use Claude Sonnet 4.6 for long-context reasoning, writing, coding, and tasks where careful instruction-following matters.
  • Use GPT-5 for general reasoning and broad knowledge tasks.
  • Use Llama for fast, cost-efficient responses on short or routine prompts.

Billing

We offer three tiers, all priced in United States dollars:

  • Free — $0 per month. 200,000 tokens. Llama 3.3 70B, GPT-5 Nano, Claude Haiku 4.5.
  • Pro — $20 per month. 5,000,000 tokens. All Pro models including Claude Sonnet 4.6 and GPT-5.
  • Ultra — $99 per month. 20,000,000 tokens. Every model including Claude Opus 4.7 and GPT-5.4 Pro.

Quota is measured in tokens. The amount that counts toward your quota for a given message equals the input tokens (your prompt and conversation context) plus the output tokens (the model’s response). If you exceed your monthly included quota on a paid plan, additional usage is billed as overage at a per-token rate marked up over the wholesale rate we are charged by our inference provider; the current overage rate is shown on your billing page before any overage is incurred.

Billing is handled by Stripe, and subscriptions auto-renew monthly. You can cancel at any time through the Customer Portal link in Billing; cancellation takes effect at the end of the current billing cycle. Refunds are not provided by default but may be issued at our discretion in cases of service unavailability or other extenuating circumstances. See the Terms of Service for the full billing policy.

Data and privacy

We do not train models on your conversations. Your prompts and responses are stored only so that we can present your conversation history back to you and so that we can operate, secure, and bill for the Service.

Chat retention depends on your tier:

  • Free: 30 days.
  • Pro: 1 year.
  • Ultra: retained indefinitely while your account is active.

You can delete individual conversations at any time from the sidebar. To delete your entire account and all associated personal information, email support@nascoder.com from your account email address; we will process the request within thirty days. A self-serve conversation export feature is on our roadmap; in the interim, please email us if you need an export.

Limits and quotas

Monthly included quotas are 200K (Free), 5M (Pro), and 20M (Ultra) tokens. Quotas reset at the start of each billing cycle. We also enforce per-minute rate limits to keep the Service responsive for everyone; if you hit a rate limit, wait a few seconds and try again.

Troubleshooting

“Model not allowed for your tier”

The model you selected is not included on your current plan. Pick a different model from the selector or upgrade in Billing.

“Quota exceeded”

You have used the included token quota for the current billing cycle. You can wait for the next cycle to start or upgrade to a higher tier in Billing.

Payment failed

Stripe was unable to charge your payment method. Update your card details through the Customer Portal in Billing. While payment is past due, paid features may be temporarily unavailable; access is restored as soon as a successful charge is recorded.

Chat hangs or never responds

Inference can occasionally stall, usually due to a transient upstream issue. Refresh the page and re-send the message. If a specific request consistently fails, switch models from the selector and try again, then email support@nascoder.com with the conversation link if the problem persists.

API access

The keys you see on the API keys page are prefixed with nas_ and are intended for use with a forthcoming nascoder command-line interface. They are not yet active; creating a key today reserves it but does not enable any programmatic access. We will announce CLI availability by email when keys go live.

Contact

For anything not covered above, email support@nascoder.com. We aim to respond within one business day. nascoder is based in the United States.