Understand the models.
Make them work.
Practical guides for running local AI, choosing models and working with developer tools. Clear steps. Linked sources.
Ollama Troubleshooting: Connection, GPU, Model and Port Checks
Start by identifying which part failed: the local server, model download, model loading, or the application connecting to Ollama. Changing all four at once…
Useful guidance starts with evidence.
We distinguish documentation, illustrative examples and recorded tests, so you know what supports each guide.
Read our editorial policy ↗Recent guides
Pin and Verify a Hugging Face Download: A Small Reproducible Example
To make a Hugging Face download repeatable, select the repository and a full commit revision, use that same revision for every file from that…
LoRA vs QLoRA: Weight Memory, Training Limits and a Fair Comparison
Short answer: LoRA trains small adapter matrices while keeping the pretrained base weights frozen. QLoRA combines adapters with a quantized frozen base to reduce…
Ollama Vision API: Send a Local Image as Base64 in Python
Short answer: In Ollama's native REST chat API, put base64-encoded image bytes in the user message's images array. A path such as ./photo.png is…
Ollama Embedding Truncation: Diagnose Long Inputs Before Indexing
Short answer: Ollama's native /api/embed endpoint defaults to truncating inputs that exceed the context window. Set truncate to false when diagnosing long-document ingestion so…
Ollama Tool Calling in Python: Validate Calls and Bound the Loop
Short answer: For Ollama's native chat API, a tool call is a proposed function name and arguments. Your application must validate it, perform the…
Ollama Empty Response? Separate Thinking from the Final Answer
Quick answer: For Ollama's native chat API, read message.content for answer text and keep message.thinking separate. An empty content field is a symptom to…