
What happened
Nano published Charakuru on GitHub; it combines Tripo and Blender, with Codex handling generation and export while the user approves images and picks among four head and body model options.
Why it matters
The split lets a user keep visual judgment while the agent does the file work, so building a 3D avatar from one illustration needs less manual Blender operation.
WHO IT HITSIndividual creators and small studios who want 3D avatars from a single illustration may find the approval-based workflow lowers the Blender skill needed, though they must supply their own Codex and Tripo accounts.
Summaries like this, in your inbox every morning.
Charakuru packages a procedure devised by developer Nano into a local confirmation app plus instructions for an AI agent. The workflow starts from a single full-body illustration: the agent draws eight images covering the head and body from four directions each, and the user can request a fix for just one image. Tripo then generates 3D models, and the user selects each head and body from four candidates. The user never lets the agent decide model candidates; visual decisions stay with the user, while the agent exports and registers intermediate files. Before completion, a walking motion lets the user check how shoulders and knees bend, and there is a step to set hair and costume sway. The finished avatar is saved in VRM 1.0 format after registering the model name, author, and usage terms. The kit runs its confirmation app locally, but image and 3D generation rely on external services, so each service's terms, fees, and credits apply separately. The license allows both non-commercial and commercial use within its conditions, but redistribution needs prior consultation, and businesses with annual revenue or cumulative funding of 100 million yen or more need a separate written license for commercial or business use.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
In a benchmark called E-Commerce Bench, 18 AI models ran up to four simulated shops for 365 days, using 18 too…

ITmedia's AI Plus released a 14-page booklet on using Gemini Notebook's (formerly NotebookLM) Studio feature…

IBM Research's ALTK-Evolve turns each customer-service agent trial into short action rules and injects them in…

A consultant who specializes in security recruiting says candidates should sign up for one of ChatGPT Plus or…

After six rounds of tuning TypeSafe AI's Jev for note selection, the author benchmarked it on 100 requests aga…

A Zenn author migrating a ten-year-old PHP/Laravel monolith — hundreds of endpoints and more than 600 shared m…
