During this period, Tsukuyomi Space received several updates: Yachiyo ' s room was added to the station ' s direct chat and voice service, the front page and navigation was changed to new features, and room settings became easier to search for. Here's a list of interesting changes and ways of doing things.
This post is based on the web interface of November 13, 2026. The chart is the actual pageshot; services, models and prices may be adjusted, which is shown on the page in real time.
I. Official AI services: start on login
The room now offers a "stop-in" chat model. When you log in using the Tsukuyomi Space account, you can directly select an open-site model without applying for or filling in a third-party API Key. The models visible on the current page include: gpt-6-luna, supports text, pictures and tools; specific optional models and capabilities are based on the settings page.
In-stations are provided for the use of the unified balance of the website and are settled in terms of actual usage. Sets the page to display the current price; the picture is automatically compressed and billed to the actual Token usage.It's a test connection.before using the balance and page tips.
How to open the station for direct chat
- Login to Tsukuyomi Space, enter Yachiyo's room.Open it again. Room Settings。
- Select in the Settings Overview Chat Model, select It's a direct supply.。
- View available balances, models and prices; if required to fill or to view bills, enter Payment centre。
- Select the available model from the model list, save the changes, and then return to the room to send the message. Connections can also be tested first, if needed, to test the accounting costs.

Figure: The site's actual room set up an overview, chat models and voice access are available here.
If you already have your own service provider key, chat model pages can still choose "API cloud" and "Ollama" for friends with local models. The configuration and cost of these approaches are based on the corresponding services, respectively, and do not provide personal services for you as a result of the selection of the station for direct supply.
Let Yachiyo talk: Official voice
Yes. Room setup, boom, voice and reading , open the Open Speech Synth, select It's a direct supply., select the official Fish Audio service, model and speech, save it and return to the room. Voice is an optional feature that can be used for regular text chats after closing.

Figure: The actual voice set page, which allows a choice between the direct supply within the station, the personal API and the in-house voice.
Page currently shows the official Fish Audio as Article 0.05 /, save and test the corresponding cost; the price and service status is based on the operating page. Speech speech is used for speech expression and does not write to chat body. If you do not want to audition for the time being, you can choose to save only settings.
Problems with use?
- No model found or no chat available: confirm that login has been made, select " stand-in direct supply " and save it, check the list of models, balances and page tips; also update the model or test the connection on the settings page.
- No sound.: Confirms that a speech synthesis has been initiated, selects the available voice services and saves them, while checking the equipment volume and browser play settings.
- Costs and billsOther Organiser Payment centre View the balance, request details and prices shown on the page. Service availability and actual settlement are based on the in-station page.
ii. Frontend UI / UX: Increased availability of places to go
The first page, which now focuses on the theme "Saving a little moonlight for Journal", brings together the entrance to the room, recent articles, ballery, encyclopedia, games and square dynamics. Enter a sentence on the first page or select the suggested topic, with which you can enter the room. The top navigator also places more easily visible access points such as reading, ballery, square, encyclopedia, games and downloads.
The rooms are re-organised into chat models, voice and reading, role and dialogue, long-term memory, appearance and layout, journal and archive, tools and extensions, Live2D debug eight categories and provide search and configuration guidance. If you want to adjust a given item, you can enter directly from the overview card or the classification navigation; the classification of settings on the mobile phone is also optimized.
Main Stage continues to maintain classification, search and sorting to facilitate switching between articles, Technology records and works; the tabulation also supports jumping to the specified page. The visual style of the page and shared navigation are harmonized, and seasonal theme and background illustrations make different regions feel more continuous.
III. Where else can we go?
- Main StageRead Announcements, Lore, Technology and Fan Works and share their work.
- A letter from Starfish.: Write a letter next to the moon-light lake, where the text meets people far away.
- Game Center: If you want to take a break, you can go check on the station.
- Yachiyo's room.: Chat with Yachiyo, adjusting the way you accompany, or continuing to explore voice and long-term memory.
Thank you for your constant company and feedback. Hopefully, this update will make it easier for friends who first came to Tsukuyomi Space to start, and for those who already know you here to find some new little surprises. The sharing of experiences and recommendations in the comment area is welcome.