AI tool comparison
Rowboat vs Sup AI
Which one should you ship with? Here is the side-by-side panel verdict, pricing read, reviewer split, and community vote comparison.
Productivity
Rowboat
AI coworker that builds a local, inspectable knowledge graph from your work
75%
Panel ship
—
Community
Free
Entry
Rowboat (YC S24) is an open-source AI coworker that connects to your email, calendar, and meeting notes, then builds a persistent knowledge graph stored as plain Markdown files on your local machine. The graph is fully inspectable — it's just a folder of .md files you can open in Obsidian, edit, or commit to git. Using this local knowledge graph, Rowboat helps draft emails in your voice, prepares meeting briefs before calls, generates docs and summaries, and answers questions about your work history. It supports MCP (Model Context Protocol) for connecting external tools like GitHub, Linear, and Notion. Runs entirely on your machine with no data sent to external servers beyond your LLM API calls. The key differentiator is transparency. Unlike AI memory systems that store knowledge in opaque vector databases or cloud embeddings, Rowboat's knowledge graph is human-readable at every step. You can audit what it knows about you, delete specific facts, and understand exactly why it drafted an email the way it did.
AI Productivity
Sup AI
Runs 339 LLMs in parallel and downweights the hallucinating ones.
50%
Panel ship
—
Community
Free
Entry
Sup AI is an ensemble AI assistant that runs your query through 339 language models simultaneously, measures per-segment confidence across all responses, and synthesizes a final answer that amplifies agreement and suppresses likely hallucinations. The team claims a 52.15% score on Humanity's Last Exam (HLE) — 7.41 percentage points above the single best model — which, if verified, would make it the highest-scoring system on the benchmark to date. The underlying mechanism works like an LLM panel: each model votes on sub-claims within the response, confidence is estimated by agreement density, and the final output surfaces high-confidence segments while flagging uncertain ones. It's designed to reduce hallucination rate on factual tasks, not improve reasoning per se — the models in the ensemble aren't doing collaborative chain-of-thought, they're voting on outputs. Sup AI was built by Ken Mueller (Stanford, CEO) and Scott Mueller (AI Research Scientist) and launched on Product Hunt today. Pricing starts with $10 in free credits, no auto-charge, with a credit card required to start. The HLE benchmark claim is the headline and will face scrutiny — if verified, this is a meaningful research result. If it's cherry-picked, it's still a usable product with a differentiated architecture.
Reviewer scorecard
“Inspectable Markdown-based memory is the right call. I can version-control the knowledge graph in git, grep through it, and actually understand what context my AI assistant has — that's more than I can say for any SaaS memory product. MCP support means it plugs into my existing toolchain.”
“The HLE claim needs independent verification, but the underlying ensemble approach is architecturally sound for factual Q&A tasks. Running 339 models is expensive — pricing will be the gating factor for production use. The $10 free credit is a fair trial.”
“Self-hosted means you're on your own for setup, sync, and maintenance. Most people using AI coworker tools want them to just work — and polished competitors like Mem.ai and Notion AI have months of production hardening. The Markdown vault is clever but also fragile at scale.”
“Extraordinary claims require extraordinary evidence. A 7.41 point jump on HLE via ensembling — without publishing methodology — smells like benchmark gaming. The latency of running 339 models in parallel is also a real concern for anything other than async research tasks.”
“Persistent, user-owned AI memory stored as plain text files is the foundation of truly personal AI assistants. When models can be swapped and knowledge graphs can be exported, you break vendor lock-in completely — Rowboat is building the right abstraction layer for the long term.”
“Model ensembling is an underexplored direction in the race to reduce hallucination. If Sup AI's approach scales, it could be more durable than fine-tuning individual models — you get the wisdom of the crowd across model families, training data, and architectures simultaneously.”
“Having an AI that actually knows my past projects, writing style, and client relationships — stored in files I control — is exactly what I've wanted. Email drafting in my own voice based on real context beats generic ChatGPT outputs every time.”
“For creative work, ensemble outputs tend to regress toward the mean — you get the most-agreed-upon version of something, which is usually the least interesting version. This is a tool for factual accuracy, not creativity. I'd stick with a single strong model for writing.”
Weekly AI Tool Verdicts
Get the next comparison in your inbox
New AI tools ship daily. We compare them before you waste an afternoon.