Framework: Model Selection Decision Tree
A decision tree for choosing the right model based on task shape, risk, latency, budget, and required structure rather than hype.
Use cases
Strategy & Planning, Development & Code
Platforms
Model-Agnostic, Tool-Agnostic
Jump to a section
Workspace-ready resource
Copy the raw material, then check the setup, output contract, and failure modes.
Decision path: 1. Is the task high-stakes or externally visible?- If yes, bias toward the most reliable model and stronger review. 2. Does the task need deep reasoning, long context, or nuanced synthesis?- If yes, favour stronger reasoning models. 3. Does the workflow need rigid structured output or tool calling?- If yes, prioritise model stability under schema constraints. 4. Is latency or cost the main constraint?- If yes, use the cheapest model that still clears the quality bar for the task. 5. Is the workflow multi-step?- If yes, use different models for different steps instead of forcing one model to do everything.Workspace translation
Turn this resource into an inspectable run.
Best next step
Use it as a review standard for the next output you save.
From resource to system
Use it once
Use the framework as a checklist, rubric, or decision aid.
Make it reusable
Attach it as project knowledge so future threads inherit the same criteria.
Decide after the run
Use it as a review standard for the next output you save.
Quality bar
Before using this, check the contract.
What input does this require?
What output should it produce?
Where can it fail?
What should a human review?
Recommended path
When to Use This
Use this when teams keep asking “which model should we use?” as if there is one correct answer across every workflow.
It is useful for builders, operators, and consultants who need a calmer model-selection process tied to task requirements.
Why It Works
The framework works because it starts from workflow constraints rather than model brand preferences.
The multi-step rule matters most. Many systems get both more reliable and cheaper once you stop using one premium model for every stage.
How to Customise
Add your own evaluation thresholds for latency, quality, and price if those matter operationally.
If your workflows depend heavily on tool use or code generation, extend the decision tree with those branches explicitly.
Limitations
This framework helps choose models sensibly, but it does not replace actual benchmarking on your own tasks.
If you do not test against real inputs, the selection will still be driven by assumptions.
Model Notes
Tool-agnostic and model-agnostic by design. The value is in the decision logic, not in any one vendor.
Related Resources
Browse FrameworksFramework: AI Tool Evaluation Matrix
A structured decision matrix for evaluating AI tools before committing. Scores tools across seven weighted criteria to cut through marketing hype and make informed choices.
Knowledge / rubric seed
Strategy & Planning · Operations & Workflow
Framework: Prompt-to-Agent Escalation Checklist
A practical checklist for deciding when a single prompt is enough and when the work genuinely needs a multi-step agent workflow.
Knowledge / rubric seed
Strategy & Planning · Operations & Workflow
Framework: AI Workflow Failure Mode Audit
A practical audit framework for diagnosing where AI workflows fail across context, tools, review, source quality, and operational design.
Knowledge / rubric seed
Operations & Workflow · Strategy & Planning
Related Guides
Need this operationalized?
Turn the pattern into a workspace system.
Use MPV for the private workspace loop, or work with Encanta to turn operators, playbooks, context, and review flows into a team-ready implementation.