How much does Teaspoon IDE cost to use?

The app itself is free and works on a BYOK basis, so you pay Google directly for what you use with no subscription or API margin. In one measured run, having AI build Othello from scratch and add a hint feature cost about $0.2 in API fees.

Can it run without sending my code to an external server?

Yes. If you choose a local LLM such as Ollama with Gemma, the only communication destination is localhost, so code does not leave the machine. There is no telemetry, and external communication is limited to requests to the LLM endpoint you configure.

Why did it change its name from Forger?

There were many projects with the same or similar names, which hurt searchability, so it was renamed to Teaspoon IDE at v0.6.0, including the package name and repository. The name 'Teaspoon' reflects the design goal of doing real work with a teaspoon of tokens.

Large Language ModelsAI Coding AssistantsAI Business & IndustryZenn AI/MLOct 6, 2026Oct 6, 2026, 10:01 JST

Teaspoon IDE hits v1.0.0, two weeks after debut

Send on LINE

3 Key Points

What happened
The developer shipped v1.0.0 of Teaspoon IDE, a standalone Electron AI coding agent integrated with file explorer, Monaco editor, AI chat, Git, and a true TTY terminal.
Why it matters
It targets business use where code must not leave the machine, since a local LLM keeps all traffic on localhost and the app is BYOK with no subscription or API markup.
What to watch
Its appeal hinges on whether BYOK and the organizational sign-in gate satisfy firms that resist monthly-subscription AI IDEs; the license bars competing commercial products until each release becomes MIT after two years.

WHO IT HITSEnterprise and contract developers who cannot send confidential code to vendor servers, since the app is free, BYOK, telemetry-free, and can run fully offline with a local LLM. Organizations needing centralized control also get a sign-in gate, virtual API keys, and budget badges.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Get news like this every morning, free

Context & Analysis

Teaspoon IDE started as an answer to concrete frustrations from using AI coding tools in contract work: not wanting confidential material sent to external servers, friction with VS Code extensions that conflict with GitHub Copilot, and the cognitive load of AI rewriting source without approval. It was renamed from Forger to Teaspoon IDE at v0.6.0, and the name encodes a token-saving design where the file tree is sent first and contents are fetched on demand.

The distance from v0.1.0 to v1.0.0 was covered in under two weeks of near-daily updates. The work split into agent reliability and robustness: handling broken command output, dealing with markdown and ::: fences, tracking failed WRITE_FILE and EDIT_FILE operations as unresolved to prevent false completion reports, distinguishing existing test failures from regressions the agent caused, and switching to a full WRITE_FILE when diff syntax keeps failing. Security work scoped file commands to the project root, blocked escapes via .. and absolute paths, and converted context to project-relative paths so OS usernames are not exposed.

For organizations, the v0.5.0 management mode adds a sign-in gate, virtual API keys, budget badges, and a model allowlist. The FSL-1.1-MIT license allows personal and business use, modification, and internal deployment, but not competing commercial products, with each release moving to MIT two years after publication. Whether this model wins over teams depends on whether BYOK and the local-LLM path satisfy procurement constraints that block subscription AI IDEs.

FAQ

How much does Teaspoon IDE cost to use?: The app itself is free and works on a BYOK basis, so you pay Google directly for what you use with no subscription or API margin. In one measured run, having AI build Othello from scratch and add a hint feature cost about $0.2 in API fees.
Can it run without sending my code to an external server?: Yes. If you choose a local LLM such as Ollama with Gemma, the only communication destination is localhost, so code does not leave the machine. There is no telemetry, and external communication is limited to requests to the LLM endpoint you configure.
Why did it change its name from Forger?: There were many projects with the same or similar names, which hurt searchability, so it was renamed to Teaspoon IDE at v0.6.0, including the package name and repository. The name 'Teaspoon' reflects the design goal of doing real work with a teaspoon of tokens.

Zenn AI/MLRead Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

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.