Open-source model
An open-source model is a machine-learning model whose relevant artifacts, such as its code or weights, are made available under terms that permit specified forms of access, use, modification, or redistribution.
LLMs in Production
A machine-learning model whose weights, code, or other defining artifacts are made available under terms permitting specified use, modification, or redistribution.
- How it is used
- Provides a starting point for tailored LLMs when training from scratch or relying on a vendor API does not fit.
- How deep it goes
- Discusses customization through finetuning and RAG, but does not specify model licenses or architecture.
- Examples
- Uses Falcon 40B as a base-model example and a pretrained coding model for a Python-focused VS Code copilot.
- What you can do after
- Apply it via a pretrained coding model and RAG to build a Python-focused VS Code copilot.
Introduction to Generative AI, Second Edition
An open-source model is a machine-learning model released under terms permitting use, modification, and redistribution.
- How it is used
- Frames the choice between accessible, customizable AI models and provider-controlled alternatives.
- How deep it goes
- Defines open source against open weights and compares practical trade-offs without explaining model implementation.
- Examples
- Uses Stable Diffusion’s community-made variants and a comparison table for open and closed language models.
- What you can do after
- Evaluate/understand whether a model is genuinely open source and weigh its access, customization, infrastructure, and safety trade-offs.
Read how Introduction to Generative AI, Second Edition explains open-source model →
Generative AI in Action
An open-source large language model is a language model whose source artifacts and permitted use are made available under an open-source license.
- How it is used
- Guides enterprise model selection by comparing open-source and commercial options for capability, licensing, integration, and governance.
- How deep it goes
- Defines the category and compares examples, licenses, capabilities, and integration paths without detailing training or inference internals.
- Examples
- Uses Llama, Dolly, Alpaca, FreedomGPT, StableCode, Hugging Face hosting, and Visual Studio Code integration as concrete examples.
- What you can do after
- Evaluate/understand model choices by matching capabilities, licenses, integration options, and enterprise requirements to a use case.
Read how Generative AI in Action explains open-source model →
The Art of AI Product Development
An open source model is a model whose source artifacts and usage rights permit users to inspect, modify, deploy, or otherwise use it under applicable licensing terms.
- How it is used
- Supports deployment and customization decisions when privacy, infrastructure control, and model economics matter.
- How deep it goes
- Explains deployment paths, model-size trade-offs, and evaluation criteria, but stops short of implementation details.
- Examples
- Uses Hugging Face, Llama, Mistral, Qwen, healthcare, finance, and smaller-versus-larger model comparisons.
- What you can do after
- Evaluate open-source candidates against quality, latency, governance, privacy, infrastructure, and deployment-cost requirements.
Read how The Art of AI Product Development explains open-source model →
Building Reliable AI Systems
A model whose weights are made available under terms that permit users to obtain and run it, and potentially modify or adapt it.
- How it is used
- Frames model hosting as a production choice involving cost, reliability, privacy, and compliance.
- How deep it goes
- Details deployment controls and infrastructure requirements, including serving, monitoring, and scaling.
- Examples
- Names LLaMA, Mistral, and Qwen, and describes locally serving and fine-tuning Mistral 7B.
- What you can do after
- Apply via Transformers and PyTorch to load a local model and generate responses.
Read how Building Reliable AI Systems explains open-source model →
Related concepts
- Llamaincludes · Llama is a family of autoregressive Transformer large language models developed by Meta and released in multiple versions and sizes.
- language modelkind of · A language model is a computational model that represents patterns in linguistic tokens to estimate their probabilities and predict or generate sequences of them.
- fine-tuningused by · Fine-tuning is adapting a pretrained machine-learning model by training it further on targeted data or objectives for a task, domain, or behavior.
- commercial modelalternative to · A commercial model is an AI model offered as a product or service under commercial terms, such as usage conditions, fees, or a proprietary license.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.