Bento/Slides, the free single-HTML-file presentation tool that reached 994 points on Hacker News on July 22, 2026, beats Gamma's $18-25/month Pro plan on speed, ownership and AI-agent editing but loses on narrative polish and export fidelity. Scored against Robert Iger's three-priorities discipline from The Ride of a Lifetime, Gamma still wins for an actual fundraising pitch; Bento wins for internal working decks.
A new presentation tool crossed 994 points on Hacker News on July 22, 2026 — high enough that three separate founders sent it to me the same afternoon. Bento/Slides is a single HTML file that is simultaneously the deck and the editor: no login, no cloud dependency, works offline, and the whole document — including live collaboration state — lives as plain JSON inside the file. It ships with an agent-editable convention: point Claude Code or another coding agent at the file with a plain-English instruction like "turn the pricing table into a chart," and the agent rewrites the JSON block directly, no export-reimport round trip. I spent two days rebuilding an actual investor deck in it against my usual tool, Gamma, to see if the hype held up.
Gamma, the incumbent I've used since 2025, now runs four tiers: Free (400 one-time AI credits, "Made with Gamma" branding), Plus at $9-12/month, Pro at $18-25/month (60 cards per prompt, custom domains, API access), and Ultra at $90-100/month for teams running heavy volume. Bento/Slides is free and open source, self-hostable, with no credit system to run out of — because there's no server-side AI generation to meter. That's the real trade: Gamma spends its subscription on AI drafting quality; Bento spends nothing because you bring your own AI agent to edit it.
I judged both against a lens most tool reviews skip: Robert Iger's discipline from The Ride of a Lifetime, specifically his rule that a leader gets three priorities, not five, and that "people will often focus on little details as a way of masking a lack of any clear, coherent, big thoughts." A pitch deck tool's real job is forcing clarity, not decorating slides. Here's how they scored on that basis, plus the practical mechanics founders actually care about:
| Criterion | Bento/Slides (free) | Gamma Pro ($18-25/mo) |
|---|---|---|
| Forces a 3-priority narrative (Iger test) | No — it's a blank canvas; sprawl is easy | Better — AI-generated outlines default to a tight structure |
| Speed from notes to first draft | Fast if you already know your content; slower for AI drafting since there's no built-in generator | Fastest — one prompt produces a full draft in minutes |
| Ownership / lock-in | Total — it's one file you keep forever, no account | None — content lives in Gamma's cloud, PPTX export can lose fidelity |
| AI-agent editing (Claude Code, Cursor) | Native — the whole doc is one editable JSON block | None — Gamma has its own AI, but no external-agent file access |
| Investor-ready polish | Functional, plain design by default | Wins — premium AI image models, custom branding, professional layouts |
The honest verdict, after actually pitching a mock version of my own deck to a colleague in both: I would not send a Bento deck to an investor cold. It's plain by default, and Iger's point about creative feedback — start with the big picture, not the details — cuts both ways here; a deck that looks unfinished invites nitpicking on formatting instead of the argument. Gamma's AI-generated structure and polish still win the actual fundraising conversation. But for the deck I build for myself before that — the one where I'm arguing with my own three priorities, cutting slides, rewriting the story five times in an afternoon — Bento is genuinely better, because I can hand it to Claude Code and say "cut this to three sections" and watch it actually happen in the file, no export cycle, no credits ticking down. I've moved my internal strategy decks to Bento and kept Gamma for anything that leaves the building.
The limits are real on both sides. Bento is one week old as a public project; there's no roadmap guarantee, no company behind it if something breaks, and the collaborative-editing feature (end-to-end encrypted, merges offline edits via an optional "blind relay") is new enough that I wouldn't trust it yet for a deck with a hard deadline. Gamma's failure mode is the opposite: you're renting the tool, the export can degrade your layout the moment you leave the platform, and at Ultra tier ($90-100/month) it's expensive for a solo founder who presents a handful of times a quarter.
Sources: Hacker News, "Show HN: Bento — An entire PowerPoint in one HTML file," July 22, 2026; Bento/Suite official site; Bento GitHub repository; Gamma official pricing page; eesel.ai, "Gamma pricing in 2026: full breakdown," June 5, 2026; Robert Iger, The Ride of a Lifetime (2019).
Related: How to Find Your Passion · Best Self-Improvement Books · How to Make Better Decisions · AI Coach App — Building It in 8 Hours