Learn AI Models and Providers
Who makes the AI models behind the apps you use, and how to tell them apart. Thirteen lessons start with the labs: OpenAI, Anthropic and Google, then the open-weight labs, then the rest of the map. Then come model names, open model licenses, and the models that make pictures, video and speech. The last lessons show where models live and run: hubs, providers and routers. It assumes nothing. It stops at who makes what and where you reach it.
Not started
0 of 13 lessons done
Level 1 · The AI labs
score0400/24 wordsThe firms that train AI models, and what each one makes. Which of them share their models with the world.
- AI Labs and Model FamiliesWhat an AI lab is, and how a lab, a model family and an app differ. Who makes GPT, Claude, Gemini, Llama and Qwen.10 min5 terms
- OpenAI, Anthropic and GoogleThe three US labs behind most closed AI models. Who makes GPT, Claude and Gemini, when each began, and how you reach them.11 min6 terms
- Llama, Mistral, DeepSeek, QwenThe four best-known open-weight families and the labs behind them. What each one shares, and what its license allows you to do.11 min5 terms
- Grok, Kimi, Cohere and NousFour more labs worth knowing, each built for a different user. Who makes Grok and Kimi, and what Cohere and Nous Research offer.10 min5 terms
- Microsoft AI, Z.ai and MiniMaxThree more labs on the map. Microsoft's own MAI models, the open GLM models from Z.ai, and the wide range MiniMax makes.9 min3 terms
Level 2 · Names, sizes and licenses
score40600/16 wordsHow to read a model's name. The open models that closed labs share, and the license that comes with each one.
- AI Model Names and SizesHow to read a model name: the family, the size tier and the version. What gpt-oss-20b, Mixtral 8x7B and o4-mini each tell you.12 min5 terms
- Gemma, Phi and gpt-ossThe small open models that Google, Microsoft and OpenAI share, though their main models stay closed. What each runs on.10 min4 terms
- Open Model LicensesWhat the license on an open model lets you do. MIT and Apache 2.0, custom licenses with limits, and fully open models like Olmo.12 min7 terms
Level 3 · Picture, video and speech models
score60750/9 wordsThe models that draw pictures, make video and turn speech into text. And the labs that make each one.
- AI Image Generation ModelsThe models that make pictures from words, such as Stable Diffusion, FLUX and GPT Image. And CLIP, which only scores a match.11 min6 terms
- AI Video and Speech ModelsVeo makes video with sound, and Whisper turns speech into text. Sora shows how a closed model can be switched off for good.10 min3 terms
Level 4 · Where models live and run
score751000/12 wordsThe sites that store models, and the firms that run them for you. And the routers that reach many models with one key.
- Hugging Face and Model HubsWhere open models are stored and shared. How to read a model page on Hugging Face, what Kaggle is for, and which uploads to trust.10 min4 terms
- AI Inference ProvidersFirms that run open models on their own chips and sell you the answers. What Groq, Together AI, Fireworks AI and Replicate each do.11 min5 terms
- OpenRouter and Model APIsThe ways to reach an AI model, from the lab's own API to one key for hundreds of models. What OpenRouter does, and what it costs.11 min3 terms
Words in this course
The 61 glossary entries this course teaches, in the order it teaches them. Open any of them on its own.
- AI Lab
- Model Parameter
- Large Language Model (LLM)
- Closed Model
- Open-Weight Model
- OpenAI
- Anthropic
- Google DeepMind
- GPT (Generative Pre-trained Transformer)
- Claude
- Gemini
- Meta AI
- Llama
- Mistral AI
- DeepSeek
- Qwen
- SpaceX Super Intelligence (SpaceXSI)
- Grok
- Moonshot AI
- Cohere
- Nous Research
- Microsoft AI
- Z.ai (Zhipu AI)
- MiniMax
- Token (AI)
- Parameter Count
- Mixtral
- OpenAI o-series
- Model Snapshot
- Gemma
- Phi
- gpt-oss
- Small Language Model (SLM)
- Software License
- Permissive License
- MIT License
- Apache License 2.0
- Allen Institute for AI (Ai2)
- Olmo
- Nemotron
- Stability AI
- Stable Diffusion
- Contrastive Language-Image Pre-Training (CLIP)
- FLUX
- GPT Image
- Nano Banana
- Veo
- Sora
- Whisper
- Model Hub
- Hugging Face
- Model Card
- Kaggle
- Inference Provider
- Groq
- Together AI
- Fireworks AI
- Replicate
- LLM API
- LLM Gateway
- OpenRouter