Companion Scout blog

DeepSeek on Janitor AI: Free or Paid? 5-Step Setup (2026)

how-tojanitor-aitips

To use DeepSeek on Janitor AI directly, create a DeepSeek API key, add a prepaid balance, then enter the full chat-completions endpoint and a current model name in Janitor’s API settings. These steps follow current provider documentation; the current Janitor field labels and successful authenticated conversation still require an in-account check. Choosing the right billing route prevents mixed keys and wallets.

There are now three distinct routes: direct DeepSeek API, Janitor+ Router and a third-party router such as OpenRouter. Do not mix their keys, model names or endpoints. Our Janitor AI proxy guide compares the three routes and their privacy trade-offs.

Key Takeaways

  • Direct setup uses https://api.deepseek.com/v1/chat/completions, not an OpenRouter URL
  • Start with deepseek-flash; choose deepseek-v4-pro when higher capability justifies the higher token rate
  • The official DeepSeek API is prepaid and metered by tokens, not a guaranteed free tier
  • Janitor+ Router is easier but requires a Janitor+ subscription, a separate Router balance and currently works on web only
  • Router credits are non-refundable except where law requires otherwise
  • If you want a simpler alternative without API setup, see our Janitor AI alternatives guide

Which DeepSeek route should you choose first?

RouteWhat you paySetupBest forFirst cost limiter
Janitor free/JLLM$0Built inTesting Janitor before spendingStandard models and about 9k context
Janitor+ / JLLM+Live subscription priceBuilt inEasier roleplay, priority and more contextSubscription renewal
Janitor+ RouterJanitor+ plus per-token walletNo external key; web onlySwitching third-party models in one placeRouter balance and model rate
Direct DeepSeek APIPrepaid DeepSeek token usageManual key + endpointCurrent DeepSeek models and direct billingInput/output tokens
OpenRouterProvider/model dependentExternal key + OpenRouter endpointModel variety and rotating free routesCredits, daily caps and provider availability

Janitor’s official subscription FAQ says Janitor+ adds five times more context, priority access and monthly swipes with its best models. Its newer Router lets subscribers browse supported third-party models, compare rates and context, and pay from a Router wallet without pasting an external key. Router is currently web-only and separate from JLLM+.

Direct DeepSeek is the right route when you specifically want DeepSeek’s current API models and billing. Router is the convenience route. OpenRouter is the variety route. JLLM+ is the low-fuss Janitor-native route.

Why connect DeepSeek directly?

Connecting your own DeepSeek API key changes this. A direct key uses DeepSeek’s documented provider endpoint and billing. Speed and availability can change; this source refresh does not include an authenticated end-to-end Janitor conversation test.

If you are comparing the setup against a hosted adult-roleplay app, start with our no-filter AI chat apps guide first. Janitor is flexible, but the “free” part depends on the model you connect and how much you chat.

DeepSeek is attractive for Janitor AI users for three reasons:

  • Price visibility: official rates are published per million input and output tokens
  • Current models: the API now exposes deepseek-flash and deepseek-v4-pro
  • Direct billing: you are not waiting for a community proxy or wondering which provider billed the request

Step 1: Create a DeepSeek Account and Get Your API Key

  1. Go to platform.deepseek.com and create an account
  2. Navigate to the API Keys section (usually under your account profile)
  3. Click “Create new API key”
  4. Name it something recognizable (e.g., “Janitor AI”)
  5. Copy the key immediately — you won’t be able to see it again

Your API key will look like: sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

Billing check: DeepSeek’s current documentation describes token billing, and Janitor’s official direct-API guide instructs users to top up a balance before creating the connection. Promotional credits can change; do not build the setup around assumed free usage.


Step 2: Get the DeepSeek API Endpoint

In Janitor AI’s Proxy URL field, use the full chat-completions endpoint:

https://api.deepseek.com/v1/chat/completions

DeepSeek’s SDK base URL is https://api.deepseek.com, but Janitor’s custom proxy field expects the full request endpoint. Mixing a base URL with a full endpoint is a common cause of 404 errors.


Step 3: Configure Janitor AI

  1. Log in to janitorai.com
  2. Click on your profile icon in the top right
  3. Select “API Settings” or “Configure API”
  4. Set API Source to Proxy or the current custom/OpenAI-compatible option
  5. In the Proxy URL field, enter: https://api.deepseek.com/v1/chat/completions
  6. In the API Key field, paste your DeepSeek key
  7. Save settings

Step 4: Select Your Model

After saving the API settings, you’ll need to specify which DeepSeek model to use. Enter the model name in the model field:

  • deepseek-flash — lower-cost starting point for most chats; supports thinking and non-thinking modes
  • deepseek-v4-pro — higher-capability and higher-cost option for more demanding scenes

Start with deepseek-flash. DeepSeek’s October 12 documentation recommends deepseek-flash. Legacy deepseek-v4-flash and deepseek-v4-flash-vision-exp aliases are still accepted, but their retired models are served by V4.1 Flash at the current Flash price. Check the provider model list rather than treating an old alias as a distinct model.


Step 5: Test the Connection

  1. Start a conversation with any character
  2. Send a simple message
  3. If you see a response within a few seconds, your connection is working
  4. If you see an error, check:
    • API key is copied correctly (no extra spaces)
    • Endpoint URL is exact
    • Your DeepSeek account has active credits

Common error: “Invalid API Key” — Double-check the key. A single character difference breaks it. Copy directly from the DeepSeek dashboard.

Common error: “Insufficient balance” or 402 — Add funds to the provider tied to the key. Janitor Router credits, DeepSeek balance and OpenRouter credits are three separate wallets.

Common error: 404 or model not found — Check that the endpoint is the full DeepSeek chat-completions URL and the model is currently listed by DeepSeek. Do not paste an OpenRouter model slug into a direct DeepSeek connection.

Common error: 429 — The provider is rate-limiting requests. Wait, reduce rerolls or switch routes; adding a Janitor+ subscription does not change a direct DeepSeek account’s provider limits.


How do you manage costs?

DeepSeek’s official table checked October 12, 2026 lists prices per million tokens:

ModelUncached input, off-peak / peakOutput, off-peak / peakCached input, off-peak / peak
deepseek-flash$0.15 / $0.30$0.60 / $1.20$0.003 / $0.006
deepseek-v4-pro$0.66 / $1.32$1.98 / $3.96$0.022 / $0.044

The vendor defines peak hours as 01:00–04:00 and 06:00–10:00 UTC, Monday–Friday, excluding Chinese public holidays. Other periods are off-peak. Rates can change.

A Flash peak-rate example: 10,000 uncached input tokens plus 2,000 output tokens costs $0.0054. This is arithmetic from documented rates, not a measured Janitor session. Long histories, output and rerolls multiply spend. Start with a small balance and verify billed usage. To estimate your own monthly bill, use the proxy cost calculator below.

Janitor+ Router shows each model’s rate in the interface and bills the same way: per token. It has no separate Router fee beyond Janitor+ and the wallet usage. Launch credits are useful for testing, expire one year after claim and are non-refundable except where required by law.


Is DeepSeek the right API option for Janitor AI?

ModelCostSpeedRoleplay Quality
DeepSeek V4.1 FlashLowest direct DeepSeek rateFast starting routeGeneral chat and roleplay
DeepSeek V4 ProHigher rate; ratio differs for input/outputCapability-firstDemanding scenes and instructions
Janitor JLLM+Included with Janitor+EasiestEveryday Janitor roleplay
Janitor+ Router modelVaries by listed modelEasy switchingThird-party models without API keys
OpenRouter modelVaries by providerFlexibleRotating free/paid model choice

There is no universal cheapest route because the prompt length and output length drive token cost. Flash is the sensible direct-DeepSeek starting point; JLLM+ is the simpler hosted route.

If token billing is the part you are trying to estimate, use the broader AI companion token-cost guide. Janitor’s API setup behaves more like a pay-per-use model than a fixed subscription.


When should you consider a different approach?

If you find the API setup process frustrating or want something that just works without configuration, you have options:

SpicyChat AI runs on managed servers with no API setup required. Adult content is available. Less flexible than Janitor AI but significantly easier to use.

Kindroid is a full companion app with its own managed infrastructure, strong memory, and encrypted conversations. If what you actually want is a companion that grows with you over time (not just roleplay sessions), Kindroid is the better answer.

See our full companion ranking if you want a broader view of what’s available.

For Janitor AI specifically, DeepSeek remains a strong low-cost API option. Use current current model names, keep the three wallets separate and choose Router only when the convenience is worth paying for both Janitor+ and tokens.

Sources & references

  • DeepSeek API documentation — current base URL, current model names and request format: Your First API Call
  • DeepSeek API documentation — current Flash/Pro peak and off-peak token rates (checked October 12, 2026): Models & Pricing
  • Historical Janitor direct-proxy guide: the formerly linked Part Two article returned 404 on October 12. Use Janitor’s current Help Center and confirm the current custom-proxy fields; this is not a freshly tested integration receipt.
  • Janitor AI Help — Janitor+ free/paid context, priority and cancellation: Subscription FAQ
  • Janitor AI Help — Router availability, wallet billing, discounts and refund limits: janitor+ Router

What will my proxy cost per message and per month?

Prices checked 2026-11-04 from each provider’s official pricing page (sources below). Estimates assume every swipe resends the full context and exclude tax. A 1,000-token reply is roughly 750 English words. Nothing you enter leaves your browser.

Proxy doctor: what does this error mean?

Paste the error your proxy shows (for example “429”, “failed to fetch” or “model not found”).

All common proxy errors
401 Unauthorized / invalid API key
Cause: The key is wrong, revoked, or pasted with a space or line break. Fix: Create a fresh key in the provider dashboard, paste it again with no spaces, and check it belongs to the same provider as the endpoint URL.
402 Payment required / insufficient balance
Cause: Your balance is empty or negative for a paid model. Fix: Top up the provider balance, or switch to a free model. On OpenRouter, a top-up of 10 credits also lifts the free-model cap from 50 to 1,000 a day.
403 Forbidden
Cause: The key does not have access to this model, the account is restricted, or the model is unavailable in your region. Fix: Check the model is enabled for your account and region in the provider dashboard, or pick another model.
404 Not found / model not found
Cause: The model ID is misspelled, retired, or the endpoint URL is wrong. Fix: Copy the exact model ID from the provider’s model list (IDs are case-sensitive), and make sure the URL ends in /chat/completions.
429 Too many requests / rate limited
Cause: You hit a per-minute or per-day cap. OpenRouter free models allow 20 a minute and 50 a day (1,000 a day after buying 10 credits). Busy upstream providers also return 429. Fix: Wait and retry, slow down swipes, try another free model, or add credit. Each swipe counts as a request.
5xx / overloaded / timeout
Cause: The upstream provider is down, overloaded or slow; it is not your settings. Fix: Retry in a few minutes or switch provider. Large contexts make timeouts more likely, so trimming chat memory helps.
“Failed to fetch” / network error
Cause: The request never reached the provider: usually a wrong endpoint URL, a typo in https, or a browser extension or network blocking it. Fix: Re-check the full endpoint URL (for example https://openrouter.ai/api/v1/chat/completions), disable blockers for the site, and try another network.
Context length exceeded
Cause: Character card + persona + memory + chat history is larger than the model or your settings allow. Fix: Lower the context size, shorten the character card or lorebook, or start a new chat with a summary.
Empty or cut-off replies
Cause: The reply-length (max tokens) setting is too low, or a reasoning model spent its tokens thinking. Fix: Raise max new tokens, or turn reasoning off for chat models that support it.
Refusals / content blocked
Cause: The model or provider’s content policy declined the request. This is policy, not a bug. Fix: Pick a provider whose published usage policy fits what you use it for. We don’t publish ways around provider safety rules.

Frequently asked questions

Can you use DeepSeek on Janitor AI?

Yes. Janitor AI supports a custom proxy/API configuration, so you can connect a DeepSeek API key and use DeepSeek models. Janitor+ subscribers can also use the web-only Router without managing an external API key.

Is DeepSeek free to use on Janitor AI?

Do not assume the official DeepSeek API is free. DeepSeek documents token-based billing and Janitor AI’s own setup guide tells users to add a prepaid balance. Free DeepSeek routes on OpenRouter can exist, but their model names and limits change.

Which DeepSeek model works best on Janitor AI?

Start with deepseek-flash for lower-cost general chat. Use deepseek-v4-pro when capability matters more than price. DeepSeek’s current API docs no longer list the old deepseek-chat and deepseek-reasoner names as the primary models.

Why is DeepSeek better than Janitor AI proxies?

A direct DeepSeek key gives you provider pricing and the newest official model names, but requires manual setup and a separate prepaid balance. Janitor+ Router removes key management; OpenRouter offers more model choice but adds another provider and its own limits.

What is Janitor+ Router?

Janitor+ Router is a web-only subscriber feature that provides third-party models inside Janitor AI without an external proxy or API key. Usage is billed per token from a Router wallet, and listed rates and context limits vary by model.