TokenStation for Linux, newest build (06b2359)

The newest build from claude/project-thread-0d2vnz, for testing. Debian/Ubuntu: curl -LO .../tokenstation_latest_amd64.deb && sudo apt install ./tokenstation_latest_amd64.deb

TokenStation5 01.004.001

Security fixes in 01.004.001

This update fixes security holes in TokenStation5's built-in Tailscale. Update if you use Tailscale to reach your PS5 (#97).

• Your tailnet's access rules now apply. Before, every device on your tailnet could reach the API, whatever the rules in your Tailscale admin console said. Now only devices your rules allow can connect, and none can until the rules arrive from Tailscale.
• A way around the API key is closed. A device on the tailnet that opened a connection and dropped it at the right moment could be let in without a key. Over Tailscale, every request now needs a key, as intended.
• HTTPS no longer trusts the leaked eMudhra certificates (Tailscale's bulletin TS-2026-013), for every connection the title makes: Tailscale, downloads and search.
• Your Tailscale sign-in key is deleted from the settings file once the PS5 has signed in, so it can't be read later over the home network's FTP.

About "Tailworm": those flaws are in Tailscale SSH and Serve in Tailscale's own tailscaled. TokenStation5 runs its own small Tailscale client, without SSH or Serve, so they don't apply to it.

Tested on a PS5 over Tailscale: a valid key works for chat, and no key, a wrong key or a deleted key is refused.

Also in this version: everything new since 01.003.000

Google's Gemma 3 12B

• Gemma 3 runs on the PS5 (#94). Gemma 3 12B (bartowski's google_gemma-3-12b-it-Q4_K_M.gguf, 6.8 GB) passed every GPU check against the CPU and 59 of 60 app-style checks, tool calls and a write-a-file agent task included. Find it in the Models page's Hugging Face search.
• Most of its layers only look back 1,024 tokens, so they keep a small rolling memory, and it fits about 29,700 tokens of context next to its weights.
• It reads prompts at 250 tokens a second and writes about 40 a second, up from 188 and 35 in its first build on the console.

Faster and steadier

• New GPU kernels for wide attention heads and sliding windows: prompts read a third faster on Gemma, and writing is 13% faster (#94).
• Switching models no longer fails with "out of GPU memory" after a bigger model was loaded: the model's memory is now split into smaller pieces (#94).
• A GPU check on the model you're serving no longer runs out of memory (#94).

Better answers out of the box

• Known-good settings (#92): each checked model (Qwen 3 8B, Qwen 2.5 Coder 7B, Llama 3.1 8B, Llama 3.2 3B, Mistral Nemo 12B and now Gemma 3 12B) gets the sampling settings that scored best on the PS5. An app's own settings still win.
• Four fixes to how tool calls and thinking are read back from the models (#92).
• Mistral Nemo 12B takes tools in its own format (#91).
• The Models page and Hugging Face search only offer models the engine can run (#90).

Your models survive updates

• The first start after a store update brings your downloaded models back, through the jailbreak's FTP server (#93). Start the title once after updating. Settings, keys and chats were already kept.

Install

Install or update from the Prospero Store, or as before: PPSA99555.zip holds the PPSA99555/ app folder.

TokenStation5 01.004.000

What's new since 01.003.000

Google's Gemma 3 12B

• Gemma 3 runs on the PS5 (#94). Gemma 3 12B (bartowski's google_gemma-3-12b-it-Q4_K_M.gguf, 6.8 GB) passed every GPU check against the CPU and 59 of 60 app-style checks, tool calls and a write-a-file agent task included. Find it in the Models page's Hugging Face search.
• Most of its layers only look back 1,024 tokens, so they keep a small rolling memory, and it fits about 29,700 tokens of context next to its weights.
• It reads prompts at 250 tokens a second and writes about 40 a second, up from 188 and 35 in its first build on the console.

Faster and steadier

• New GPU kernels for wide attention heads and sliding windows: prompts read a third faster on Gemma, and writing is 13% faster (#94).
• Switching models no longer fails with "out of GPU memory" after a bigger model was loaded: the model's memory is now split into smaller pieces (#94).
• A GPU check on the model you're serving no longer runs out of memory (#94).

Better answers out of the box

• Known-good settings (#92): each checked model (Qwen 3 8B, Qwen 2.5 Coder 7B, Llama 3.1 8B, Llama 3.2 3B, Mistral Nemo 12B and now Gemma 3 12B) gets the sampling settings that scored best on the PS5. An app's own settings still win.
• Four fixes to how tool calls and thinking are read back from the models (#92).
• Mistral Nemo 12B takes tools in its own format (#91).
• The Models page and Hugging Face search only offer models the engine can run (#90).

Your models survive updates

• The first start after a store update brings your downloaded models back, through the jailbreak's FTP server (#93). Start the title once after updating. Settings, keys and chats were already kept.

Install

Install or update from the Prospero Store, or as before: PPSA99555.zip holds the PPSA99555/ app folder.

TokenStation5 01.003.000

TokenStation5 01.003.000

TokenStation5 01.002.000

What's new since 01.001.000

Instructions in Settings

• Settings > Instructions opens a list of guides: Set up an app, Use it away from home (Tailscale, no open ports) and Add models from a computer (FTP) (#73, #77).

Models and downloads

• Up to three downloads at once, and unfinished downloads carry on after the app restarts (#74).
• The TV no longer freezes while a download saves; the detailed log no longer slows searches (#74).
• Sizes are counted the way the PS5 does (1,024-based), so the model card reads "of the GPU's 11.4 GB" (#74).
• Large downloads inside the app are still limited by the PS5 after the first few hundred MB; copying a model from a computer over FTP is much faster (see the new guide).

Your data

• Settings, API keys, chats and usage are kept in the app's download area as well, and put back after a store update (#75). Models are not kept: a store update or an uninstall removes them, so keep a copy of big models on your computer.

Store listing

• New store description and author name (#72).

Install

Install or update from the Prospero Store, or as before: PPSA99555.zip holds the PPSA99555/ app folder.

TokenStation5 01.001.000

What's new since 01.000.001

Chat

• Saved chats on the Chat page: a sidebar with New chat, Search chats and your chat history, newest first. Reopen a chat and carry on where you left off; remove one with square twice (#69, #70).
• Thinking models (Qwen3) show their thinking apart from the answer (#59).

Models

• Search Hugging Face from the Models page for models that fit this PS5, and download them on the console. Chat (Instruct) models are listed first; others are marked "May not chat well" (#58, #63).
• Download or remove a model, the speech model and the voice right on the console (#47, #48).
• Models now get the largest context that fits (e.g. 44,032 tokens for Llama 3.1 8B, 32,768 for Qwen2.5 Coder 7B and Qwen3 8B).
• Llama 3.1 and 3.2 now handle long conversations correctly (#67).

Apps (Kilo Code, Continue, Cline, Open WebUI…)

• API keys: make a key per app in Settings, and optionally require one (#57).
• Set up an app guide in Settings with the address, model name and context window to use (#57).
• Compatibility fixes for coding tools: tool calls in the forms models actually write, max_tokens: -1, chunked requests, a clear message when an app uses https:// by mistake, and a proper error when a prompt is too long (#56, #59, #60, #66).

Faster and quieter

• Hugging Face searches take ~11 s instead of minutes; the log no longer holds anything up (#61, #62, #65).
• Detailed log switch in Settings (off by default) for troubleshooting downloads and searches (#64).

Install

Install as before: PPSA99555.zip holds the PPSA99555/ app folder. Models, settings, keys and saved chats live in the same folder and are not part of the zip.

TokenStation5 01.000.001

First release of Token Station 5.