Skip to content

NanoTeams 1.8.0 — Ollama

Choose a tag to compare

View all tags
@jmstajim jmstajim released this 27 Jul 19:57
· 12 commits to main since this release
1.8.0
52ed6ad

NanoTeams runs on Ollama as well as LM Studio. Pick the provider in Settings → LLM → Server; Ollama defaults to http://127.0.0.1:11434. Teams, tools, Vision, delegation and the Autovisor behave the same on both.

✨ Ollama

  • Per-surface providers — per-role overrides, Vision, the Autovisor model, the Bash judge and Generate Team each pick a provider or inherit the global one.
  • Model Details card — state, context length, quantization, family, capabilities; on Ollama also VRAM and keep-alive expiry.
  • Downloaded Models card — lists what's on disk with sizes and removes it.

🛠 Improved

  • Planning phase — a role reads the work folder with read-only tools and records what it found before implementing. Toggle in the role editor's General tab; on by default for the Software Engineer.
  • Prompt cache misses are visible — a CACHE ×N pill in the status bar, broken down by cause and by caller. A miss costs a full re-prefill: ~350 ms warm against ~4.3–6.1 s cold at 13k tokens.
  • Models are released at run end, on task deletion and on team switch. Duplicate instances are reaped. A model a running step is using is not unloaded.

🐛 Fixed

  • A role that finished just before you quit could be demoted at next launch: the task read Working indefinitely and every review control disappeared. Affected tasks repair themselves on start.
  • Run now on the Autovisor did nothing while the manager was mid-pass or waiting for input.
  • Switching the Autovisor off left its tab showing the manager's chat, with no way to re-enable from that screen.

System Requirements

  • macOS 15.0 or later
  • Ollama 0.9 or later, or LM Studio 0.4.0 or later
  • Apple Silicon recommended for best local-LLM performance
  • macOS 26+ required for private voice dictation

Install

  1. Download NanoTeams.app.zip and extract it.
  2. Drag NanoTeams.app into /Applications.
  3. If macOS blocks the app on first launch: System Settings → Privacy & Security → Open Anyway.
  4. Start Ollama or LM Studio, load a model, then launch NanoTeams and pick your provider in Settings → LLM.

Build from Source

git clone https://github.com/jmstajim/NanoTeams.git
cd NanoTeams
xcodebuild -project NanoTeams.xcodeproj -scheme NanoTeams -configuration Release build

Links

Topics: macOS · AI coding assistant · local LLM · Ollama · LM Studio · multi-agent AI · AI agents · on-device · offline AI · private AI · automation · open source · Apple Silicon · Swift · SwiftUI · agentic chat · code generation

Assets 3
Loading