Selecting the Optimal LLM(s)
Description
This capability focuses on choosing the best LLM or set of models from a shortlist based on structured evaluation, business goals, and operational fit. It involves comparing final results, weighing tradeoffs, and aligning on a go-forward decision.
Why it's Important
Choosing the wrong model can lead to lower performance, higher costs, or poor user trust. An effective selection process ensures decisions are grounded in evidence, not opinion. It enables teams to move forward with confidence, knowing that the chosen LLM balances speed, quality, cost, and safety in the context of their specific goals.
Why it's Challenging @ Scale
- Tradeoffs are rarely clear-cut: A model that performs well on quality may lag on cost, latency, or control.
- Final scores can be close: Teams often face difficult calls between similarly performing options.
- Stakeholders may disagree: Preferences around vendors, architectures, or features can create tension.
- Context matters deeply: The “best” model may vary by workflow, geography, or integration path.
- Lack of process introduces risk: Without a repeatable decision framework, choices may feel arbitrary or political.
Complexity
Medium to High: Maturing this capability requires a well-aligned scoring rubric, stakeholder input, documented rationale, and repeatable decision criteria that reflect both technical and business priorities.
Taking Action
Though most organizations begin their GenAI journey with significant knowledge gaps, there are targeted actions that can be taken to accelerate the process. Select your group’s current maturity, based on your assessment results, and act today.
Exploring
Experimenting
- Explore Key Concepts & Best Practices: Complete the Evaluating and Selecting the Best Model(s) for Your GenAI Solution workshop (2 hrs.) to understand foundational key concepts and explore applied best practices
Click here to review Specific Areas of Focus
- Outlining the Model Evaluation Lifecycle
- Understanding Model Types and Capabilities
- Aligning Evaluation to Solution Objectives
- Comparing Commercial vs. Open Source Options
- Establishing a Reusable Evaluation Framework
- Define Your Action Plan: Outline concrete, prioritized steps your organization will take to implement GenAI Strategy
Click here to review Specific Areas of Focus
- Align on your Current State and define your Target State
- Create an actionable enablement plan
- Define target timeline and measures of success
- Deliver Quick Wins: Small, high-impact GenAI projects that can demonstrate tangible value in a short time frame
Click here to review Specific Areas of Focus
- Summarize Top Model Scores: Consolidate evaluation results across core metrics in a visual, shareable format.
- Draft a Selection Brief: Document pros, cons, and tradeoffs for the top 2-3 candidate models.
- Run a Decision Workshop: Convene key stakeholders to review findings and align on selection criteria.
Experimenting
Lifting-Off
- Complete one or more of our Deep Dive Courses: Begin exploring key concepts and best practices, including:
Click here to review Specific Areas of Focus
- Defining Your Model Objectives & Requirements
- Model Evaluation Data Assessment and Prep
- Selecting In-Scope Models
- LLM Evaluation
- Nail It Before You Scale It: Assess and optimize your solution or process before adopting it at scale
Click here to review Specific Areas of Focus
- Assess Your Proposed Solution or Process: Confirm that final selection decisions are grounded in data, not preference.
- Define in-scope Processes and Guardrails: Establish decision-making frameworks and thresholds for each model attribute.
- Close any Data or Measurement Gaps: Fill in missing test results, usage feedback, or cost data to inform final selection.
- Define Your Adoption & Scaling Plan: Create a structured roadmap for how GenAI solutions will be rolled out across teams, workflows, or business units
Click here to review Specific Areas of Focus
- Define Your Phased Implementation Plan: Choose which models to deploy where and when based on use case maturity.
- Build Awareness and Finalize Enablers: Communicate the rationale behind selected models and provide usage guidance.
- Operationalize Your Comms Plan: Share the decision logic across business, technical, and governance stakeholders.
Lifting-Off
Accelerating
- Formalize Your Best Practices: Document and standardize what’s working to ensure consistent, scalable success across teams and use cases
Click here to review Specific Areas of Focus
- Create a Model Selection Framework: Define how evaluation results, business priorities, and operational constraints are weighted.
- Publish Model Selection Templates: Provide a standard slide or document format to support selection briefings.
- Store Selection Rationale for Future Review: Keep a searchable archive of past decisions and the reasoning behind them.
- Accelerate Your Adoption: Intensify efforts to embed GenAI across your organization by expanding use cases, increasing user engagement, and removing adoption barriers
Click here to review Specific Areas of Focus
- Share Approved Model Lists by Use Case: Help teams move faster by offering vetted options for specific task types.
- Enable Cross-Functional Buy-In: Include legal, compliance, and product teams in the final review and endorsement process.
- Reduce Time from Evaluation to Deployment: Streamline the handoff from model selection to solution buildout.
- Celebrate Your Wins: Publicly acknowledge team accomplishments to build and sustain adoption momentum
Click here to review Specific Areas of Focus
- Highlight a Model Selection Win: Share where selecting the right LLM led to faster delivery or better outcomes.
- Recognize Contributors to the Process: Acknowledge those who helped interpret data, shape frameworks, or lead decisions.
- Publish Before-and-After Model Results: Show the difference in quality or efficiency enabled by the final selection.
Accelerating
Breaking-Away
- Streamline & Embed: Integrate GenAI into core workflows while eliminating friction points to make usage seamless and routine
Click here to review Specific Areas of Focus
- Automate the Selection Workflow: Build tools that score, compare, and recommend models based on live evaluation data.
- Pre-Approve Models for Common Needs: Reduce friction by offering default choices with prebuilt governance and integration assets.
- Track Model Decisions Across Teams: Centralize model selection outcomes for visibility, reuse, and risk management.
- Leverage Automation: Using GenAI-powered tools and workflows to streamline repetitive tasks, enhance operational efficiency, and reduce manual effort
Click here to review Specific Areas of Focus
- Auto-Generate Selection Briefs: Summarize key tradeoffs and metrics using scripts or LLMs.
- Flag Selection Biases Automatically: Detect when non-scoring factors may be skewing final decisions.
- Use Decision Trees or Rulesets for Selection: Recommend models based on predefined logic and business rules.
- Evolve & Further Accelerate: Continuously refine GenAI strategies based on insights and outcomes, while expanding into more complex or high-impact use cases
Click here to review Specific Areas of Focus
- Analyze Selection Trends Over Time: Review patterns in past decisions to improve future guidance and frameworks.
- Expand to Multi-Model Strategies: Choose multiple models for fallback, routing, or use-case-specific optimization.
- Tie Selection to Business Impact: Track how model choices affect performance, satisfaction, and ROI across teams.
Key "Watchouts"
As you take action you’ll want to avoid:
- Choosing based on hype or preference: Avoid selecting models based solely on familiarity or marketing claims.
- Overlooking downstream costs or constraints: A high-performing model may still be hard to deploy or govern.
- Failing to document decisions: Without clear rationale, model selections are hard to explain or revise later.
- Misaligning model to use case needs: Even the best general model may underperform in specialized workflows.
- Neglecting future flexibility: Locking in a model too early can limit options as needs or tools evolve.
Targeted Benefits
While Selecting the Optimal LLM(s) can be challenging, its benefits are clear and compelling, including:
- Faster go-to-market: Confident decisions reduce delay between evaluation and solution buildout.
- Stronger cross-functional alignment: Transparent processes improve buy-in across technical and business teams.
- Improved model outcomes: Selecting the right LLM increases performance, usability, and trust.
- Lower risk: Structured selection reduces chances of choosing models with hidden costs or weaknesses.
- Repeatable success: Standardizing how models are selected enables scale, consistency, and reuse.