Model selection wizard

LLM Selection Decision Tree

Use this interactive decision tree to find the most suitable large language model (LLM) based on your enterprise use case, budget constraints, and compliance requirements.

Choosing the right large language model (LLM) involves balancing business use case demands, budget constraints, and compliance obligations. This wizard guides you through these factors to suggest candidates best aligned with your needs.

Answer the questions below regarding your intended use, expected scale, required compliance standards, and available budget to receive tailored model recommendations.

Inputs

Primary enterprise use case

Select the closest description matching your AI application.

Expected monthly usage volume

Approximate number of API calls or interactions per month.

Compliance requirements

Choose all relevant compliance or security standards you must meet.

Monthly budget for LLM usage (USD)

Provide your monthly budget dedicated to LLM costs (API or license fees).

Result

Model Recommendation

function(inputs) { const { use_case, expected_usage_volume, compliance_requirements, budget_monthly } = inputs; const models = [ { name: 'OpenAI GPT-4 API', costPer1000: 0.03, supports: ['all'], compliances: ['gdpr', 'soc2'], maxVolume: 'high' }, { name: 'Anthropic Claude 3', costPer1000: 0.045, supports: ['customer_support','content_creation','research','virtual_assistant'], compliances: ['soc2'], maxVolume: 'medium' }, { name: 'Cohere Command R', costPer1000: 0.012, supports: ['data_analysis','research'], compliances: ['gdpr'], maxVolume: 'high' }, { name: 'Google PaLM 2 Enterprise', costPer1000: 0.05, supports: ['all'], compliances: ['hipaa','gdpr','fedramp','soc2'], maxVolume: 'high' }, { name: 'Meta LLaMa 2 Enterprise', costPer1000: 0.025, supports: ['all'], compliances: ['gdpr'], maxVolume: 'medium' }]; let filtered = models.filter(m => (m.supports.includes(use_case) || m.supports.includes('all')) && (compliance_requirements === 'none' || m.compliances.includes(compliance_requirements)) && (m.maxVolume === expected_usage_volume || (m.maxVolume === 'high' && expected_usage_volume !== 'high') || (m.maxVolume === 'medium' && expected_usage_volume === 'low'))); filtered = filtered.filter(m => { let maxCalls = expected_usage_volume === 'low' ? 10000 : (expected_usage_volume === 'medium' ? 100000 : 500000); let costEstimate = (maxCalls/1000)*m.costPer1000; return costEstimate <= budget_monthly; }); if(filtered.length === 0) return 'No LLM found matching all criteria within budget.'; filtered.sort((a,b) => { let aCost = (expected_usage_volume === 'low' ? 10000 : (expected_usage_volume === 'medium' ? 100000 : 500000))/1000*a.costPer1000; let bCost = (expected_usage_volume === 'low' ? 10000 : (expected_usage_volume === 'medium' ? 100000 : 500000))/1000*b.costPer1000; return aCost - bCost; }); return 'Recommended model: ' + filtered[0].name + '. Estimated monthly cost approx. $' + ((expected_usage_volume === 'low' ? 10000 : (expected_usage_volume === 'medium' ? 100000 : 500000))/1000*filtered[0].costPer1000).toFixed(2) + '.'; }

—

Your recommended LLM

Note

This recommendation is based on publicly available pricing and compliance claims as of mid-2024. Actual costs and compliance responsibilities require direct vendor confirmation and contract review.

Subsequent sections unlock after submit