Types of AI Models

Last updated on 2026-08-27 | Edit this page

Overview

Questions

  • What types of AI models are available for research?
  • What are the computational requirements for different model types?
  • How do open-source models compare to proprietary ones?

Objectives

  • Distinguish between LLMs, vision models, and scientific ML models
  • Understand model sizes, quantization, and GPU requirements
  • Compare open-source models available on Sagehen HPC with cloud alternatives

Large Language Models in Detail

Open-Weight Models for Sagehen HPC

Specific models turn over quickly. Choose along three axes that do not: licence, size, and whether the download is gated.

The examples below are current as of August 2026.

gpt-oss (OpenAI) — gpt-oss-20b, gpt-oss-120b

  • Apache 2.0, ungated: pip install and go, no access request
  • 20b runs in about 16 GB; 120b fits a single A100 (80 GB)
  • The path of least resistance for workshop work and for publication

Qwen3 (Alibaba)

  • Apache 2.0, ungated, unusually wide range of sizes
  • Strong multilingual coverage — worth knowing if your corpus is not English

Mistral Small (Mistral AI)

  • Apache 2.0; efficient general-purpose model that fits a single L40S

Phi-4-mini (Microsoft)

  • MIT licence, about 3.8B parameters
  • The smallest thing here that is still useful; good for prototyping a pipeline before committing GPU hours

Gemma 3 (Google) — multimodal, handles image input. Note the licence is Google’s own Gemma terms, not an OSI-approved open-source licence.

Llama 4 (Meta) — mixture-of-experts; Scout is 109B total / 17B active. Requires accepting the Llama Community Licence before download.

A decision starting from whether you can use an AI tool, branching on the data tier. PUBLIC data may use any tool, hosted or local. PROPRIETARY data may use local models on Sagehen HPC, or a hosted one with written approval. RESTRICTED data may use local models only and never a hosted service. A note advises classifying the data before opening the tool rather than after pasting something in.
Classify first; the tier narrows the options for you.
Callout

Prefer Ungated Models Unless You Need Otherwise

A gated model (Llama, among others) means: accept a licence on the model’s page, generate a token, then authenticate before downloading. That is fine for sustained project work, but it stalls a workshop and adds a licence term you must honour at publication.

Unless a gated model gives you something you specifically need, start with an Apache 2.0 or MIT one.

Quantization: Making Models Smaller

Quantization reduces model precision to use less GPU memory:

  • 32-bit (full precision): Best quality, most memory
  • 16-bit (half precision): Negligible quality loss, half the memory
  • 4-bit quantization: 75% memory reduction, slight quality trade-off

For many research tasks (summarization, classification), the difference between 16-bit and 4-bit is negligible. Always test on your actual data.

Vision and Multimodal Models

Model Task VRAM Notes
YOLO Object detection 4-8GB Fast, real-time capable
Stable Diffusion 1.5 Image generation 4GB Text-to-image
Stable Diffusion XL Image generation 24GB Higher quality
CLIP Image-text matching 4GB Zero-shot classification
SAM (Meta) Image segmentation 8-16GB Segment anything

Scientific ML Models

Protein science: AlphaFold, ESMFold – predict 3D protein structure from amino acid sequences. Often require 20-100GB+ memory.

Climate and earth science: FourCastNet, climate downscaling models, satellite imagery analysis. Require massive datasets and sometimes distributed training across multiple GPUs.

Genomics: Language models trained on protein/RNA sequences, disease prediction models. Often require specialized data preprocessing.

Choosing the Right Model

Callout

Start Small, Scale Up

Begin with the smallest model that meets your quality needs:

  1. Try Phi-4-mini or an 8B-class model such as Qwen3 8B first
  2. If quality is insufficient, scale to a 20-30B model such as gpt-oss-20b
  3. Use a 100B-class model only when smaller ones clearly underperform
  4. Always use 4-bit quantization unless you need full precision

This conserves GPU resources and reduces queue wait times on Sagehen.

Challenge

Challenge: Model Selection for Research Tasks

For each scenario, recommend a model type and size:

  1. Summarizing 50 published research papers
  2. Classifying sentiment in 10,000 customer reviews (proprietary data)
  3. Detecting objects in microscopy images
  4. Generating Python code to analyze a public dataset
  1. Summarizing papers (public data): Cloud AI is fine since papers are public. If you prefer local: an 8B-class model on an L40S.
  2. Sentiment classification (proprietary): Local AI required. A fine-tuned encoder such as BERT, or an 8B-class instruct model, on an L40S. Batch processing is efficient.
  3. Object detection in microscopy: YOLO or ResNet on L40S depending on image resolution and batch size.
  4. Code generation (public data): Cloud AI is acceptable. Locally, gpt-oss-20b on an L40S gives good code generation results.

Note that only scenario 2’s answer was forced by the data. The others were choices about cost and convenience — which is the normal case.

Key Points
  • Useful local LLMs range from about 4B to over 100B parameters
  • Choose on licence, size and whether the download is gated – not on leaderboards
  • Quantization (4-bit) reduces memory by 75% with minimal quality loss
  • Vision models (YOLO, Stable Diffusion) and scientific ML have specialized uses
  • Start with the smallest model that meets your quality threshold and scale up