What it is
omakase is a CLI + browser tool that pulls your AniList or MyAnimeList history, combines it with a short taste profile you write in markdown, and asks any LLM — local or cloud — to recommend 5–10 anime tied to your scoring patterns. It supports nine LLM backends out of the box (Ollama, LM Studio, OpenAI, Anthropic, Gemini, DeepSeek, OpenRouter, Groq, Together).
There are two ways to run it. The public demo at omakase.jhinx.dev is bring-your-own-key and public-safe — paste a key, get a tasting menu, nothing is stored. The private Plus tier at anime.jhinx.dev is in testing with multi-tenant accounts, AniList OAuth write-back so picks land directly in your Plan to Watch, and an Overseerr request flow so a recommendation can become a download in one step.
Why I built it
Streaming platforms and tracker sites all optimize for engagement. AniList's "recommendations" tab is the same popular shows everyone already watches. I scored 200+ anime, rated 17 of them a 9 or 10, and the platform still recommended Demon Slayer.
The thesis: an LLM that gets to read both your written taste and your scoring patterns can do far better than collaborative filtering — especially in a niche with deep history and strong personal aesthetics like anime. The wine sommelier metaphor is the right one. Omakase = "I'll leave it to the chef."
What I learned
- LLM JSON mode quality varies wildly across providers — Anthropic and DeepSeek's reasoning models need looser parsing, OpenAI/Gemini/most others honor
response_formatcleanly. The codebase has asupports_jsonflag per preset so it falls back gracefully. - A 40-line OpenAI-compatible client covers 6 of 9 backends. Anthropic and Gemini each need their own ~50-line client because their wire formats diverge. Ollama is its own thing.
- Self-hosted UI matters more than I expected. Most people don't want to install a CLI just to try a recommender — the
omakase webcommand was a late add that turned out to be the primary entry point.
Status & next steps
The public demo runs at omakase.jhinx.dev; the private Plus tier is in testing at anime.jhinx.dev with AniList write-back and the Overseerr request flow under review. Next up: smarter season-aware filtering (don't recommend an anime that's a sequel to something already in your dropped list), and deciding when Plus is ready beyond a single account.