Model Selection
AWS LabsOfficialPublished by AWS Labs
- Skill
- Research and analysis
- Data and AI
- Apache-2.0
- Low risk
- Has scripts
AutoGineer summary
Written by AutoGineer from the skill’s own files, not by the author
The model-selection skill streamlines the process of choosing a base model for AI development. It interacts with the user to identify their requirements, queries available SageMaker Hubs to list compatible models, and provides transparency regarding model licenses and performance benchmarks. By ensuring the exact Hub model ID is resolved, it prepares the user for subsequent development or fine-tuning tasks.
What it does
- Verifies the active AWS region for SageMaker access.
- Discovers and lists available SageMaker Hubs for model retrieval.
- Retrieves and presents a list of available models with their associated licenses.
- Cross-references model options with provided benchmark data.
- Confirms the final model selection with the user to ensure alignment with their goals.
When to use it
- When you need to select or change your base model for a project.
- When you are unsure which model best fits your specific use case.
- When you need to resolve a specific Hub model ID for a known model family.
- When you want to evaluate base models before proceeding to fine-tuning.
Advantages
- Ensures model selection is based on available resources in the user's specific AWS region.
- Provides clear visibility into model licensing requirements.
- Integrates benchmark data to support informed decision-making.
- Standardizes the model selection process to prevent configuration errors.
Usage examples
- “I need to pick a base model for my new project, what are my options?”
- The agent will check your AWS region, query the available SageMaker Hubs, and present a list of models with their licenses and benchmark context to help you decide.
- “I want to use a Llama model, can you help me set that up?”
- The agent will identify the available Llama models in your Hub, display their specific license details, and ask you to confirm the exact model ID to proceed.
Install
Download the skill as a ZIP:
https://mp-staging.autogineer.com/agent-skills/awslabs/model-selection/downloadUnzip it into the cross-client skills folder of your project:
mkdir -p .agents/skills && unzip model-selection.zip -d .agents/skills
Documentation by AWS Labs
Model Selection
Guides the user through selecting a base model based on their use case.
When to Use
User asks which model to use
User wants to select or change their base model
User mentions a model name or family (e.g., "Llama", "Mistral", "Nova") — the exact Hub model ID still needs to be resolved
User wants to evaluate a base model before deciding whether to finetune
Prerequisites
A
use_case_spec.mdfile exists. If not, activate the use-case-specification skill to generate it first.
Workflow
Step 1: Check Region
Run:
python -c "import boto3; print(boto3.session.Session().region_name)"None→ STOP. Tell user: "Set your region viaexport AWS_DEFAULT_REGION=us-west-2oraws configure."Set → store REGION in context, continue.
Step 2: Discover Hub
List all available SageMaker Hubs in the user's region by calling the SageMaker
ListHubsAPI using theaws___call_awstool.From the results, filter out any hub whose
HubDescriptioncontains "AI Registry" — these do not contain JumpStart models.The remaining hubs are eligible (e.g.,
SageMakerPublicHuband any private hubs).If exactly one eligible hub exists, use it automatically — do not ask the user.
If multiple eligible hubs exist, present them to the user and ask which one to use. Example:
I found the following model hubs: - SageMakerPublicHub — SageMaker Public Hub - Private-Hub-XYZ — Private Hub models Which hub would you like to use?Store the selected hub name for use in subsequent steps.
Step 3: Select Base Model
First, retrieve all available SageMaker Hub model names by running: python model-selection/scripts/get_model_names.py <hub-name>.
Present all available models to the user with their licenses before making any recommendations. Cross-reference the model list with references/model-licenses.md and display each as <model name> - [<license>](<url>). For example: "Qwen3-4B - Apache 2.0"
If you already know the model the user wants to use (from conversation context or planning files), confirm that it's in the list, display its license, and move on. Otherwise, help the user pick a model following the instructions in references/model-selection.md.
Important: Make sure to remember this list of available models when helping with model selection. Don't recommend a model that's not available to the user.
Step 4: Confirm Selection
Present a summary to the user:
Here's what we've selected:
- Base model: [model name]Ask if they'd like to proceed with this model.
References
references/model-selection.md— Model selection instructions and benchmark descriptionsreferences/model-licenses.md— Model license information for display during model selection
Files
- ATTRIBUTION.mdAdded by AutoGineer914 B
- LICENSELicence file added by AutoGineer9.9 KiB
- NOTICELicence file added by AutoGineer89 B
- SKILL.md3.3 KiB
- references/benchmarks/agenticIndex.md2.8 KiB
- references/benchmarks/codingIndex.md2.8 KiB
- references/benchmarks/gpqa.md2.8 KiB
- references/benchmarks/hle.md2.8 KiB
- references/benchmarks/ifbench.md2.7 KiB
- references/benchmarks/intelligenceIndex.md2.7 KiB
- references/benchmarks/mmmuPro.md3.0 KiB
- references/benchmarks/tau2.md2.8 KiB
- references/model-licenses.md7.8 KiB
- references/model-selection.md8.0 KiB
- scripts/get_model_names.pyScript1.0 KiB
references/benchmarks/mmmuPro.md
MMMU-Pro
College-level multimodal (vision+text) questions across six academic disciplines. Only scored for multimodal models.
Use this for: Tasks combining visual and textual inputs — charts, diagrams, visual reasoning.
Source: Artificial Analysis (artificialanalysis.ai), June 2026. "—" = no data. For multimodal models without scores, infer from similar models in the same family, but tell the user you're inferring. Text-only models cannot be scored on this benchmark — do not infer scores for them.
| # | Model | Family | Score |
|---|---|---|---|
| 1 | Qwen3.5 27B (mode: reasoning) | Qwen | 75.0% |
| 2 | Qwen3.6 27B (mode: reasoning) | Qwen | 74.6% |
| 3 | Gemma 4 31B (mode: reasoning) | 73.4% | |
| 4 | Qwen3.5 9B (mode: reasoning) | Qwen | 69.2% |
| 5 | Qwen3.5 4B (mode: reasoning) | Qwen | 65.4% |
| 6 | Nova 2.0 Lite (mode: high) | Amazon Nova | 63.8% |
| 7 | Llama 4 Scout 17B | Meta Llama | 52.9% |
| 8 | Nova Pro | Amazon Nova | 44.3% |
| 9 | Nova Lite | Amazon Nova | 37.8% |
| — | Llama 3.3 70B Instruct | Meta Llama | — |
| — | Llama 3.2 3B Instruct | Meta Llama | — |
| — | Llama 3.2 1B Instruct | Meta Llama | — |
| — | Qwen3 32B (mode: reasoning) | Qwen | — |
| — | Qwen3 14B (mode: reasoning) | Qwen | — |
| — | Qwen3 8B (mode: reasoning) | Qwen | — |
| — | Qwen3 4B (mode: reasoning) | Qwen | — |
| — | Qwen3 1.7B (mode: reasoning) | Qwen | — |
| — | Qwen3 0.6B (mode: reasoning) | Qwen | — |
| — | Qwen2.5 72B Instruct | Qwen | — |
| — | Qwen2.5 32B Instruct | Qwen | — |
| — | DeepSeek R1 Distill Llama 70B | DeepSeek | — |
| — | DeepSeek R1 Distill Llama 8B | DeepSeek | — |
| — | DeepSeek R1 Distill Qwen 32B | DeepSeek | — |
| — | DeepSeek R1 Distill Qwen 14B | DeepSeek | — |
| — | DeepSeek R1 Distill Qwen 1.5B | DeepSeek | — |
| — | GPT-OSS 120B (mode: medium (averaged)) | OpenAI | — |
| — | GPT-OSS 20B (mode: medium (averaged)) | OpenAI | — |
| — | Nova Micro | Amazon Nova | — |
| — | Nemotron 3 Nano 30B | NVIDIA | — |
| — | Llama 3.1 8B Instruct | Meta Llama | — |
| — | Qwen2.5 14B Instruct | Qwen | — |
| — | Qwen2.5 7B Instruct | Qwen | — |
| — | DeepSeek R1 Distill Qwen 7B | DeepSeek | — |
Requirements and compatibility
Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta
“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.
Needs: Needs Python
Security
What our scanner noticed· Low risk
- Scripts that your agent may run
This skill includes scripts that your agent may run. Read them before you use it.
A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.
Attribution and licence
- Published by
- AWS Labs
- Publisher
- AWS Labs (Official)
- Source
- awslabs/agent-plugins · plugins/sagemaker-ai/skills/model-selection · commit fbefb37
- Licence
- Apache License 2.0 (redistribution allowed)
- LICENSE (added by AutoGineer from LICENSE)
- NOTICE (added by AutoGineer from NOTICE)
- ATTRIBUTION.md (written by AutoGineer)
- Synced
AutoGineer is not affiliated with or endorsed by AWS Labs.
Download ZIP
The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download model-selection.zip