Unquant

Opening your workspace…

Unquant

How to use Unquant as a Janitor AI proxy

Janitor AI can send your chats to any OpenAI-compatible API through its proxy setting. Point it at Unquant and every message is unlimited, at one flat monthly price.

Setup guide2 min readUpdated 1 October 2026
  1. 1

    Open API Settings

    Open any chat on Janitor AI, open the menu and choose API Settings.

  2. 2

    Fill in the proxy

    Select the Proxy option (shown as a custom or "other API" choice), then enter:

    Proxy
    Model nameglm-5.3-flash
    or, on a Qwen planqwen-3.8-flash-next
    Proxy URLhttps://api.unquant.io/v1/chat/completions
    API keyYour Unquant API key
  3. 3

    Tune the generation settings

    Generation settings
    Context sizeup to 262,144 on GLM; 65,536 to 262,144 on Qwen by tier
    Max new tokens600 to 1,500 for long-form replies
    Temperature0.7 to 1.0 for creative writing

Which model for Janitor AI?

Qwen 3.8 Flash Next

Long roleplay at the lowest priceQuick and long-context, from $49/month

GLM-5.3-Flash

Plot-heavy campaignsStronger reasoning and consistency, from $109/month

Both are uncensored at the tensor level, so there is no jailbreak prompt to maintain.

Troubleshooting

404
Network or "not found" error

The proxy URL must end in /v1/chat/completions.

401
A valid API key is required

Paste the full key again; revoked or mistyped keys are rejected.

403
Not included in your subscription

The model ID doesn't match your plan: GLM plans use glm-5.3-flash, Qwen plans qwen-3.8-flash-next.

429
Concurrent stream limit reached

Another generation is still running on your account. Let it finish, then retry.

Tip
Settings ignored

Save, then reload the chat page.

Keep reading