Smartloop

Frontier quality. Local First — run models locally on your device to extract information, generate new content, create skills, and connect to external systems.

[Smartloop @smartloop]

🚀 Smartloop Studio 1.0.62 is here!

✨ Upgrade to Pro right in the app with Mac App Store in-app purchase
💳 Top up your credits without leaving Studio
🎨 Easier tab management for a cleaner workspace

#localAI #SLM #macOS

[Mehfuz Hossain 🚀 @mehfuzh]

Frontier output, a fraction of the tokens.

A yes/no decision doesn't need a frontier model. Neither does checking whether a document is relevant. @smartloop sends each part of a request only as far as it needs to go.

01 Split the request
A request becomes small typed steps: collect, judge, extract, tool, fetch, reason. Each step gets only the context it needs. No step gets the whole conversation.

02 Spend tokens where they count
Decisions run on the fast System 1 model, capped at 64 tokens. Tool resolution gets room to reason. Only content steps load a bigger capability model.

03 Filter before you prompt
Retrieved context is judged for relevance before it enters a prompt. An unrelated document costs zero tokens.

04 Share weights, skip the bill
One model host keeps weights loaded across every project. If a model is skipped, the stream says why.

Here is what local first means. If your machine has the resources and the model, the step runs on your machine. If it doesn't, the step is delegated to a cloud model. Open and frontier models run as one system.

On real hardware:
Contract review on an M4 Mac, 24GB.
Orchestration, hybrid search, model selection and eval on a five-year-old Linux machine running Qwen.

Built in Rust on llama.cpp. Quantized open models on macOS, Linux and Windows. MCP tools, reusable skills, agents, OpenAI-compatible API. Studio 1.0.58 added Legal and Healthcare templates, including contract review, plus one-click MCP connections.

It started as private document processing for European financial institutions. Embedding every document into a single model hit its limits, so I split the work into steps. That was two years ago, mostly nights and weekends.

I'm looking for a few enterprise design partners in privacy-sensitive industries, starting with finance and law. Bring a real workflow and we'll build around it.

#LocalAI

https://t.co/AQL0KLXE4S

[Smartloop @smartloop]

🚀 Smartloop Studio 1.0.62 is here!

💳 Upgrade to Pro in-app, Mac App Store included
🔍 Monthly web search credits shown in Settings
🐛 Long answers no longer stall a chat
📚 Local API reference at /docs

Frontier quality. Local first.

#localAI #privacy #LLM

[Smartloop @smartloop]

Frontier quality. Local first.

A fast System 1 model plans every request and routes each step to the open model built for it. Decisions get a tight budget. Irrelevant context is filtered before it costs a token. No per-token API bill.

https://t.co/7RSvmP0GvY

#LocalAI

[Smartloop @smartloop]

@verbove https://t.co/z2UtroMQly

[Smartloop @smartloop]

Processing contracts locally on an M4 Mac with 24GB. No cloud, nothing uploaded.

Smartloop runs /contract-review over the Halbrook Wren LLP sample (five fictional contracts) and flags the discrepancies between them.

Try the same sample: https://t.co/JlIwmzJzfp

#localAI

[Smartloop @smartloop]

Hybrid AI search on terminal #docent #localAI

[Smartloop @smartloop]

What's under the hood in 1.0.58:

🦀 Runtime ported to Rust
🧠 Vision & capability steps go to the largest model that fits
💾 Models stay loaded; idle ones unload to free memory
🌐 URL fetching built in, no extra MCP needed

[Smartloop @smartloop]

🚀 Smartloop Studio 1.0.58 is here!

⚖️ New Legal & Healthcare templates, incl. contract review
🔌 One-click MCP connections: Trello, Google Analytics & more
⚡ Runtime rewritten in Rust, with smarter model loading

Frontier quality. Local first.

#localAI #privacy #LLM

[Mehfuz Hossain 🚀 @mehfuzh]

Frontier quality. Local First @smartloop

https://t.co/frcfLGD8BF

[Smartloop @smartloop]

docent - A tiny open-source tool (written in rust) for your terminal to do AI search and more #LocalAI

https://t.co/FZQ3MJEe2k

[Mehfuz Hossain 🚀 @mehfuzh]

If you're using doc-to-chat wrappers and sending your personal and legal documents to a product you don't know, do yourself a favor: download @smartloop from the App Store, pick a #Qwen model from the list, add your documents, set up your MCP and skills, and optionally enable AI search.

All of this costs $0 (unless you need AI search <=$5), and your data stays with you. #DocToChat

[Smartloop @smartloop]

🦀 Docent: open-source terminal app, 100% Rust.

📄 Process your personal documents
🌐 Cross-validate with the latest web info
🔒 Runs 100% on your device

Your Claude Code for documents.

💻 macOS · Windows · Linux · DGX Spark
🎥 Demo: M4 Mac, 24GB
⬇️ https://t.co/rkfNAl1JQd

[Smartloop @smartloop]

Thanks for putting docent on #pitchhut

https://t.co/8yE6Qqi6uN

[Smartloop @smartloop]

🚀 Smartloop Studio 1.0.54 is here!

🗂️ Drag tabs to reorder them
⚡ Chats keep running in the background — work in two tabs across two projects at once
🎯 Picks the lightest model that can do the job

Private, local AI on your machine.

#localAI #privacy #LLM

[Smartloop @smartloop]

🚀 Smartloop CLI 1.0.26 is here!

📊 /usage: searches left, tokens &amp; cost saved
⚠️ Heads-up before you run out of search credits
✨ /upgrade: Pro gets 1,000 searches a month
⌨️ ctrl+s toggles web search

Private, local AI on your machine.

#localAI #privacy #LLM

[Smartloop @smartloop]

The new Smartloop TUI, inspired by Claude Code from @claudeai. Use it side by side with Claude Code: hand it your document processing (attachments) and AI search. It runs 100% locally, so those tokens cost you nothing. Save those for your coding with Claude Code. #LocalAI