Tested answer: DeepSeek worked through Janitor AI’s Proxy provider in one dated private-character test using deepseek-v4-flash. Nine positive checks passed, including grounding, same-chat recall, a fresh-chat control, settings persistence, and visible incremental rendering. The later recovery gate did not pass, so this guide reports that limitation instead of presenting the run as fully complete.
What the August 2026 test established
- Positive checks: J01 through J09 passed their frozen expected outputs.
- Tested route: Janitor’s Proxy provider with the DeepSeek API base and Janitor’s visible option to add the Chat Completions path.
- Tested model:
deepseek-v4-flash, which was also present in the dated direct model inventory. - Invalid-key control: Janitor displayed a visible authentication message ending in
(401). - Recovery gate: R01 failed strict matching because the response contained two extra lines before the required recovery marker.
- Not run: the missing-endpoint, invalid-model, their recovery controls, and the final revoked-key check.
Evidence boundary: This was one dated test in a third-party interface. It does not establish an official Janitor AI–DeepSeek partnership, Janitor’s internal routing or key-storage design, a universal privacy guarantee, or support for every account and future interface version.
Tested DeepSeek settings for Janitor AI
| Field | Dated tested value | What it means |
|---|---|---|
| Provider | Proxy | The Janitor route visibly selected for this test |
| Proxy URL | https://api.deepseek.com | The official DeepSeek API base used in the saved configuration |
| Path behavior | Janitor’s Add /chat/completions behavior | The interface completed the Chat Completions route; do not duplicate the path |
| API key | Dedicated temporary DeepSeek key | Entered privately and never included in publishable evidence |
| Model | deepseek-v4-flash | The exact model ID tested on August 11, 2026 |
| Custom prompt | Blank | No test facts were injected through the proxy prompt |
Janitor AI can change labels and field behavior. Before entering a key, confirm that the current panel still separates the base URL from the optional Chat Completions path. DeepSeek documents both an OpenAI-compatible API base and a Chat Completions endpoint, but the third-party interface decides what its own field expects.
What you need before setup
- A signed-in Janitor AI account that currently exposes the Proxy provider.
- A DeepSeek Platform account, a task-specific API key, and usable API balance.
- A private synthetic character for the first test. Avoid real names, customer data, or confidential conversations.
- The current model list from DeepSeek rather than a model name copied from an older tutorial.
Use the DeepSeek API key guide for creation and rotation. A reusable production key should not be used for a tutorial or included in a screenshot, recording, support message, or public character.
How to connect DeepSeek to Janitor AI
1. Create a dedicated DeepSeek API key
Create a separate key for the Janitor configuration so it can be rotated without disrupting another application. Copy it directly into Janitor’s API key field. Do not paste it into the character definition or chat.
2. Check the live DeepSeek model inventory
DeepSeek provides an official Models endpoint. The August 11 preflight returned deepseek-v4-flash and deepseek-v4-pro; the Janitor UI test used deepseek-v4-flash. Availability can change, so use a currently returned ID.
3. Create a new Proxy configuration
In the tested Janitor chat, open Settings, select Proxy, and create a new configuration. Give it a clear name that identifies DeepSeek without including the key or account details.
4. Enter the base URL, key, and model
Enter https://api.deepseek.com as the Proxy URL, use Janitor’s visible option to add /chat/completions, enter the dedicated key, select deepseek-v4-flash, and leave the proxy custom prompt blank. If the current interface requests a full endpoint instead, follow its visible field contract rather than combining both forms.
5. Save, reload, and confirm the active provider
Save the configuration and reload once if Janitor instructs you to do so. The dated test showed a using proxy badge after selection. That badge proves only the visible Janitor state; it does not reveal the complete upstream route.
6. Send one small connection prompt
Start with a short deterministic prompt instead of a long roleplay message. In J01, the frozen request asked for two exact lines and the response matched both lines. This established that the saved configuration produced a model response in that chat.
Original Janitor AI and DeepSeek test results
The test used a private fictional character named Avery Flint, fixed synthetic facts, serial prompts, and exact-output grading. No answer was regenerated to obtain a preferred result.
| ID | Check | Observed result | Grade |
|---|---|---|---|
| J01 | Connection | Returned exactly CONNECTION_OK and MODEL_RESPONSE_READY | Pass |
| J02 | Character grounding | Returned the exact name, callsign, opening time, and drawer color as one JSON object | Pass |
| J03 | Store chat-local facts | Returned exactly SAVED | Pass |
| J04 | Scenario retrieval | Returned exactly mint tea | Pass |
| J05 | Multi-turn recall | Returned the three supplied chat facts in valid one-line JSON and the required key order | Pass |
| J06 | Same-chat reload | Returned the same three facts after one operator-attested reload | Pass |
| J07 | Fresh-chat control | Returned exactly NO_PRIOR_CHAT_DATA in one operator-attested fresh chat | Pass |
| J08 | Settings persistence | Returned exactly SETTINGS-PERSISTENCE-08 after one operator-attested reload | Pass |
| J09 | Visible rendering | Returned the exact 12-line sequence; two partial UI states appeared before completion | Pass |
| N01 | Invalid-key control | Displayed the recorded authentication message with visible status 401 | Pass |
| R01 | Known-good recovery | Returned two extra lines before RECOVERY-AUTH-01 | Fail |
| N02 onward | Remaining negative and recovery cases | Not run because the protocol required a stop after a failed recovery gate | Not run |
Interpretation limits: J06 and J08 rely on contemporaneous operator attestations that a reload occurred; a still image cannot prove the reload by itself. J07 describes one fresh-chat observation, not a platform-wide isolation or security guarantee. The positive checks show what the model returned under this fixture, not how Janitor stores or routes information internally.
What the visible streaming check showed
In the 21-second J09 recording, the first partial response state was visible at about 9.00 seconds, a second distinct partial state appeared at about 9.25 seconds, and the complete 12-line response was visible at about 9.50 seconds. This supports a narrow statement: incremental rendering was visible in the Janitor UI during this dated run. It does not prove wire-level server-sent events, DeepSeek chunk boundaries, or a direct browser-to-provider route.
Invalid API key result and the stopped recovery gate
N01 used a separate configuration with a synthetic invalid key while preserving the known-good configuration. Janitor displayed this exact message:
Your AI provider rejected the API key or account permissions. Update the credentials in Settings and try again. (401)
This is direct evidence of the visible Janitor response in N01. It should not be generalized to every Janitor authentication failure, and it does not disclose anything about Janitor’s internal handling of credentials.
After the operator switched back to the preserved known-good configuration, R01 required the response to be only RECOVERY-AUTH-01. Instead, the visible response contained three lines:
CONNECTION_OK
MODEL_RESPONSE_READY
RECOVERY-AUTH-01Because two extra lines appeared, R01 failed its exact-output rule and the protocol stopped. One possible explanation is that earlier lines remained queued or were replayed, but that is a hypothesis rather than an established cause. We did not selectively rerun R01, and we did not run N02, R02, N03, R03, or N04.
Fix common DeepSeek errors in Janitor AI
| Visible problem | Safe action | Evidence status |
|---|---|---|
Authentication message with (401) | Confirm that the key belongs to DeepSeek, is active, and was entered without extra characters. Re-enter it once; do not loop. | Observed in N01 with a deliberately invalid synthetic key |
| Wrong endpoint or duplicated path | Check whether the current Janitor field expects a base URL with path-appending enabled or a full endpoint, but not both. | Not deliberately tested in this stopped run |
| Invalid or retired model name | Use the official Models endpoint and copy a currently returned ID. | Not deliberately tested in this stopped run |
| Balance or rate-limit error | Check DeepSeek Platform billing or reduce request frequency. Do not run a load test through Janitor. | Not deliberately triggered |
| Endless loading | Stop after a reasonable timeout, preserve the visible state, and test a minimal request directly before changing multiple fields. | Not established by this run |
DeepSeek’s official error reference explains provider-side meanings. Janitor may summarize or rewrite an upstream response, so diagnose from the exact visible text and change one setting at a time. The site’s broader DeepSeek error-code guide covers generic API troubleshooting.
Which DeepSeek model should you use?
The integration test used deepseek-v4-flash. That proves compatibility only for this dated setup check; it is not a roleplay-quality ranking. Before configuring Janitor, query the current model inventory and choose an ID that is present there. The DeepSeek V4 model guide owns the broader model comparison.
Thinking mode and advanced controls
DeepSeek documents thinking behavior and model-specific reasoning controls, but the Janitor test did not prove the complete outbound request or a specific thinking setting. Do not infer that a hidden parameter was sent merely because the model displayed a thoughts control. Use DeepSeek’s thinking-mode documentation for the provider contract and describe Janitor only from visible controls.
How much does DeepSeek cost through Janitor AI?
DeepSeek API usage is billed according to provider rates and token usage. The separate direct-provider preflight used 36 cache-miss input tokens and 25 output tokens, with a rate-times-usage calculation of $0.00001204. That tiny direct-control amount is not the total Janitor test cost because a complete Janitor usage delta was not established.
Conversation history, output length, model selection, thinking behavior, and caching can materially change cost. Use the DeepSeek API cost calculator for your own workload and recheck the current official pricing page.
API key and privacy checklist
- Use a dedicated key and rotate it when the integration is no longer needed.
- Never publish the key, a masked prefix, a private chat URL, or an account identifier.
- Assume messages and character instructions can be sent to the provider configured in Janitor.
- Use a private synthetic character for troubleshooting.
- Do not paste personal, customer, medical, legal, financial, or confidential information into a test.
- Do not infer encryption, retention, storage, or routing guarantees from a successful response.
The broad DeepSeek integrations hub explains integration patterns. The generic DeepSeek API guide covers the request contract outside Janitor.
What this test did not establish
We did not establish Janitor’s internal key storage, encryption, provider routing, long-term retention, mobile behavior, public-character behavior, moderation behavior, rate-limit thresholds, long-context capacity, or roleplay superiority. The negative matrix stopped after R01, so this page does not claim tested Janitor behavior for a missing endpoint, invalid model, revoked key, or recovery after those cases. This limitation follows the site’s independent DeepSeek AI guide policy of separating observations, documentation, hypotheses, and unknowns.
Frequently asked questions
Does DeepSeek work with Janitor AI?
Yes, in one dated private test Janitor’s Proxy provider returned valid responses from a configuration using the DeepSeek API base and deepseek-v4-flash. Nine positive checks passed. The later strict recovery control failed because the response contained extra lines, so the complete negative-test matrix was not finished.
What DeepSeek URL should I enter in Janitor AI?
The August 11, 2026 test used https://api.deepseek.com as the Proxy URL with Janitor’s visible option to add /chat/completions. Check the current field labels before saving so you do not append the path twice.
Which DeepSeek model name should I use in Janitor AI?
Use an ID returned by DeepSeek’s current Models endpoint. The dated Janitor test used deepseek-v4-flash; this is a compatibility observation, not a universal quality recommendation.
Is OpenRouter required to use DeepSeek on Janitor AI?
No. This test used a direct DeepSeek API configuration. If you use another relay, its endpoint, key, and model slug must all belong to that relay; do not mix credentials or model IDs between providers.
Why does Janitor AI show a network or API error with DeepSeek?
Check the exact visible text, then verify the provider, URL form, API key, model ID, and balance one at a time. In the controlled invalid-key case, Janitor displayed an authentication message ending in (401). Other negative cases were not run in this stopped protocol.
Does Janitor AI expose DeepSeek thinking mode?
The dated test did not prove a specific thinking parameter in the outbound request. Treat only the current visible Janitor controls as observed and use DeepSeek’s official thinking-mode guide for the provider-side contract.
Is DeepSeek free to use through Janitor AI?
No permanent free-use claim is supported. DeepSeek API requests are billed under the provider’s current rates and your token usage. Janitor-inclusive cost was not fully measured in this run.
Can I still use deepseek-chat or deepseek-reasoner in Janitor AI?
Do not assume an older alias remains valid. Query DeepSeek’s current Models endpoint and use a model ID it returns. The tested ID in this guide was deepseek-v4-flash.
Official sources
- DeepSeek API quick start
- DeepSeek Models endpoint
- DeepSeek Chat Completions
- DeepSeek model pricing
- DeepSeek thinking mode
- DeepSeek error codes
- DeepSeek rate limits
Original UI test: August 11, 2026. Official-documentation snapshot: August 11, 2026. The guide reports the stopped protocol as observed and does not convert unrun checks into results.