Local LLMs with Ollama on macOS

A developer documented a working setup for running local LLMs with Ollama on a 24GB MacBook Pro M4 Pro, storing models on an external drive via the OLLAMA_MODELS environment variable. The writeup covers terminal, menu-bar app, symlink and LaunchAgent methods for persisting the path, plus fixes for port conflicts and unmounted drives. On the test machine, gpt-oss:20b loaded from the external drive and generated at roughly 52 tok/s under Ollama 0.40.14.

Setup for a MacBook Pro M4 Pro 24GB with models stored on the external drive /Volumes/ROG PRO . Verified on this machine: gpt-oss:20b loads from the drive and generates at ~52 tok/s Ollama 0.40.14 . brew install ollama CLI or download the app from https://ollama.com Check: ollama -v Models live in OLLAMA MODELS . Default is ~/.ollama/models . mkdir -p /Volumes/ROG PRO/ollama/models touch /Volumes/ROG PRO/ollama/models/.t && rm /Volumes/ROG PRO/ollama/models/.t writable? If not writable: Finder → drive → Cmd+I → turn off "Ignore ownership". NTFS drives are read-only on macOS; use APFS. | Setup | Command | |---|---| | Terminal ollama serve | echo 'export OLLAMA MODELS=/Volumes/ROG PRO/ollama/models' ~/.zshrc && source ~/.zshrc | | Menu bar app until reboot | launchctl setenv OLLAMA MODELS /Volumes/ROG PRO/ollama/models , then quit and reopen the app | | Symlink works everywhere | mv ~/.ollama/models /Volumes/ROG PRO/ollama/ ; ln -sfn /Volumes/ROG PRO/ollama/models ~/.ollama/models | Persistent menu-bar setting: create ~/Library/LaunchAgents/ollama.models.plist : <?xml version="1.0" encoding="UTF-8"? < DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd" <plist version="1.0" <dict <key Label</key <string ollama.models</string <key ProgramArguments</key <array <string /bin/launchctl</string <string setenv</string <string OLLAMA MODELS</string <string /Volumes/ROG PRO/ollama/models</string </array <key RunAtLoad</key <true/ </dict </plist Load it: launchctl load ~/Library/LaunchAgents/ollama.models.plist - "address already in use" on ollama serve : an instance is already running menu bar app or brew service and ignores your env var. Stop it first: pkill -f ollama; brew services stop ollama 2 /dev/null; lsof -i :11434 should print nothing - Drive unplugged : Ollama can't see your models or creates an empty folder. Mount the drive before starting it. - macOS may ask for Removable Volumes / Full Disk Access for Ollama. Allow it. OLLAMA MODELS=/Volumes/ROG PRO/ollama/models ollama serve Leave it running; use a second terminal for everything else. Server: http://127.0.0.1:11434 . Useful server variables: | Variable | Purpose | Example | |---|---|---| | OLLAMA MODELS | Model storage path | /Volumes/ROG PRO/ollama/models | | OLLAMA CONTEXT LENGTH | Default context window default 4096 | 32768 | | OLLAMA KEEP ALIVE | How long a model stays in RAM default 5m | 30m | | OLLAMA MAX LOADED MODELS | Concurrent models | 1 | | OLLAMA HOST | Bind address | 127.0.0.1:11434 | ollama pull gpt-oss:20b download ollama list installed models ollama run gpt-oss:20b interactive chat /bye to exit ollama run gpt-oss:20b "text" one-shot ollama ps loaded models ollama stop gpt-oss:20b unload from RAM ollama show gpt-oss:20b details ollama cp a b copy/rename ollama rm gpt-oss:20b delete frees drive space Verify storage: ls /Volumes/ROG PRO/ollama/models/blobs and du -sh /Volumes/ROG PRO/ollama/models . Budget about 17–20GB for the model. Q4 K M costs ~0.6 GB per billion parameters. Speeds below are third-party estimates modelfit.io , not benchmarks. Only gpt-oss:20b has been measured here ~52 tok/s . | Model | Size | Est. speed | Use | |---|---|---|---| | gpt-oss:20b | ~14 GB | ~53 tok/s | Daily driver, reasoning, coding | | qwen3.5:9b | ~7 GB | ~43 tok/s | Multimodal, light on RAM | | gemma4:26b | ~16 GB | ~53 tok/s | Text + images | | lfm2:24b-a2b | ~14 GB | ~92 tok/s | Agents / tool calling | | qwen3.8:27b | ~17.5 GB | ~14 tok/s | Max quality, tight fit | | ministral-3:14b | ~9 GB | ~24 tok/s | Multimodal 14B | Coding: gpt-oss:20b is the easy choice in Ollama. JetBrains Mellum2.1 12B-A2.5B Thinking is coding-specific but has no Ollama tag see §8 . Avoid: 70B models ~43GB, won't fit , Q8 above 14B, 32B at Q4 swaps . Docker: docker run -d -p 3000:8080 \ -e OLLAMA BASE URL=http://host.docker.internal:11434 \ -v open-webui:/app/backend/data \ --name open-webui --restart always \ ghcr.io/open-webui/open-webui:main Without Docker Python 3.11 : pip install open-webui && open-webui serve port 8080 . Open http://localhost:3000 , create the local admin account, pick the model. If none appear: Settings → Connections → Ollama URL http://host.docker.internal:11434 Docker or http://localhost:11434 . Open WebUI's own data chats, settings is on the internal disk, not ROG PRO. brew install sst/tap/opencode or: npm i -g opencode-ai Raise the context first. The 4096 default is too small for agent prompts: OLLAMA CONTEXT LENGTH=32768 OLLAMA MODELS=/Volumes/ROG PRO/ollama/models ollama serve 32K fits comfortably with gpt-oss:20b . Drop to 16K if memory gets tight. ~/.config/opencode/opencode.json global or opencode.json in the project: { "$schema": "https://opencode.ai/config.json", "provider": { "ollama": { "npm": "@ai-sdk/openai-compatible", "name": "Ollama local ", "options": { "baseURL": "http://localhost:11434/v1" }, "models": { "gpt-oss:20b": { "name": "GPT-OSS 20B" } } } } } Run opencode in a project directory, then /models and choose Ollama local . A 20B local model handles small edits and questions. It is weaker than hosted models at long multi-step agent work, so review its changes. - LM Studio : My Models tab → change models directory to the drive. - llama.cpp e.g. Mellum2.1 : LLAMA CACHE=/Volumes/ROG PRO/llama llama-server -hf JetBrains/Mellum2.1-12B-A2.5B-Thinking-GGUF:Q4 K M - Hugging Face / MLX : export HF HOME=/Volumes/ROG PRO/hf Native curl http://localhost:11434/api/chat -d '{"model":"gpt-oss:20b","messages": {"role":"user","content":"hi"} ,"stream":false}' OpenAI-compatible what OpenCode and most tools use curl http://localhost:11434/v1/chat/completions -H 'Content-Type: application/json' \ -d '{"model":"gpt-oss:20b","messages": {"role":"user","content":"hi"} }' | Symptom | Fix | |---|---| | Models missing after reboot | OLLAMA MODELS not persisted, or drive not mounted. See §2. | | Pull writes to ~/.ollama | Server started without the env var. Restart per §2 gotchas. | | Slow or swapping | Model too big, or too many apps open. Use a smaller model, lower context, ollama stop . | | Agent forgets or truncates | Context still 4096. Set OLLAMA CONTEXT LENGTH . | | Open WebUI sees no models | Wrong Ollama URL. Use host.docker.internal from Docker. | | Slow first response | Initial model load ~18s here . Raise OLLAMA KEEP ALIVE . |