Jev-style System One models answer typed questions with a probability per option instead of text. What runs locally, on what hardware, and how to set it up.

Decision models — start herehub Run Jev-style decision models locally with Ollama (Nimble, Tev1)Open Jev reproductions beyond Ollama: Winnow, Decider, JevK5, AutoJev, Laya Decision-model confidence, thresholds and calibration — use the probabilities safely Decision models on images: classify receipts, screenshots and forms locally with Clef How to evaluate a decision model on your own data Decision models on phones, Jetson and laptops

Troubleshooting39 files →

Paste the error you are seeing. Grouped by operating system, because that is what decides the fix.

Windows · 22CUDA out of memory Ollama: "model requires more system memory than is available"Model loads but runs painfully slow LM Studio: "Failed to load model" (insufficient memory)"No CUDA-capable device is detected"Out of memory at long context (the KV cache, not the weights)AMD GPU not detected / ROCm unsupported (Ollama & llama.cpp)Which GGUF quant should I download? (Q4 vs Q5 vs Q8)Out of disk space / failed GGUF download for large models Local LLM repeating itself or stuck in a loop Ollama "connection refused"Ollama /v1/systemone errors: 400, 404, 413 explained Ollama /v1/systemone image errors: 400, 413, HEIC and sideways photos Local LLM outputs gibberish? Wrong chat template Ollama runs on the CPU on Windows despite an NVIDIA GPU CUDA works on Windows but not inside WSL2 Windows shared GPU memory makes your model crawl, not crash "ollama is not recognized as an internal or external command"Move your models off C:Can't reach Ollama from another device Your laptop runs the model on the integrated GPU, not the RTX SmartScreen and antivirus are blocking your local AI install

Linux · 23CUDA out of memory Ollama: "model requires more system memory than is available"Model loads but runs painfully slow LM Studio: "Failed to load model" (insufficient memory)"No CUDA-capable device is detected"Out of memory at long context (the KV cache, not the weights)AMD GPU not detected / ROCm unsupported (Ollama & llama.cpp)Which GGUF quant should I download? (Q4 vs Q5 vs Q8)Out of disk space / failed GGUF download for large models Local LLM repeating itself or stuck in a loop Ollama "connection refused"Ollama /v1/systemone errors: 400, 404, 413 explained Ollama /v1/systemone image errors: 400, 413, HEIC and sideways photos Local LLM outputs gibberish? Wrong chat template CUDA works on Windows but not inside WSL2 nvidia-smi stopped working after a kernel update "Failed to initialize NVML: Driver/library version mismatch"Secure Boot is blocking your GPU driver module Your Ollama environment variables are ignored on Linux Your Docker container can't see the GPU ROCm doesn't support your AMD card "Permission denied" on /dev/kfd Ollama on a headless Linux server

Apple · 20Ollama: "model requires more system memory than is available"Model loads but runs painfully slow LM Studio: "Failed to load model" (insufficient memory)Apple Silicon: model won't use the GPU / Metal not engaged Out of memory at long context (the KV cache, not the weights)Which GGUF quant should I download? (Q4 vs Q5 vs Q8)Out of disk space / failed GGUF download for large models Local LLM repeating itself or stuck in a loop Ollama "connection refused"Ollama /v1/systemone errors: 400, 404, 413 explained Ollama /v1/systemone image errors: 400, 413, HEIC and sideways photos Local LLM outputs gibberish? Wrong chat template Your Mac is swapping to disk instead of running the model Raising the Metal VRAM ceiling on Apple Silicon "App is damaged and can't be opened"Running local LLMs on an Intel Mac MLX won't install, or runs on the CPU Your MacBook is fast for two minutes, then half as fast The model crashes your iPhone app, or drops to a smaller one iPhone model downloads that fail, fill your storage, or stop