Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.monostate.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
Choosing Your Training Approach

Not every AI problem needs the same solution. Sometimes you need full training, sometimes just fine-tuning, and sometimes no training at all.

​
The Three Approaches

Prompt Engineering

No training neededUse existing models with clever prompts
  • Instant results
  • Zero training cost
  • 0-10 examples needed

Fine-Tuning

Recommended approachAdapt pre-trained models to your needs
  • Consistent behavior
  • 100-10K examples
  • Hours to days

Training from Scratch

Rarely neededBuild completely new models
  • Full control
  • Millions of examples
  • Very expensive

​
1. Prompt Engineering (No Training)

Use existing models with carefully crafted prompts. What it is: Writing instructions that get the model to do what you want without any training. Example: 
You are a customer service agent for TechCorp.
Always be polite and helpful.
Product return policy: 30 days with receipt.

Customer: Can I return my laptop?
When to use:
  • Testing ideas quickly
  • Few examples available
  • Task is within model’s capabilities
  • Budget/time constraints
Pros:
  • Zero training time
  • No training data needed
  • Instant results
  • Free to try
Cons:
  • Limited customization
  • Inconsistent results
  • Can’t add new knowledge
  • Higher inference cost
Adapt a pre-trained model to your specific needs. What it is: Taking a model that already understands language/images and teaching it your specific task. When to use:
  • Have hundreds to thousands of examples
  • Need consistent behavior
  • Want to add domain knowledge
  • Building production systems
Pros:
  • Faster than training from scratch
  • Needs less data
  • Better performance than prompting
  • Lower inference cost than prompting
Cons:
  • Requires training data
  • Needs compute resources
  • Takes time to train
  • Can overfit on small datasets

​
3. Training from Scratch

Build a completely new model. What it is: Starting with random weights and training on massive datasets. When to use:
  • Creating foundational models
  • Completely novel architectures
  • Unlimited data and compute
  • Research purposes
Pros:
  • Full control
  • Can create novel capabilities
  • No inherited biases
Cons:
  • Needs massive data (millions of examples)
  • Extremely expensive
  • Takes weeks to months
  • Usually unnecessary

​
Decision Framework

​
Quick Decision Guide

1

Check your data

Do you have training examples?
  • No → Go with Prompt Engineering
  • Yes → Continue to Step 2
2

Count your examples

How many examples do you have?
  • Less than 100 → Use Prompt Engineering
  • 100-10,000 → Perfect for Fine-tuning
  • Millions → Could train from scratch (but why?)
3

Evaluate your needs

What’s most important?
  • Speed to deploy → Prompt Engineering
  • Consistent behavior → Fine-tuning
  • Novel architecture → Training from scratch

​
Detailed Comparison

AspectPrompt EngineeringFine-TuningTraining from Scratch
Data Needed0-10 examples100-10,000 examplesMillions of examples
Time to DeployMinutesHours to daysWeeks to months
Cost$0 upfront$10-1,000$10,000+
CustomizationLimitedHighComplete
New KnowledgeNoYesYes
ConsistencyVariableHighHigh
MaintenanceUpdate promptsRetrain periodicallyContinuous training

​
Approach by Use Case

When to use Prompt Engineering:
  • General FAQs
  • Simple routing
  • Low volume
  • Testing phase
When to use Fine-Tuning:
  • Company knowledge
  • Brand voice
  • High volume
  • Complex products

​
Fine-Tuning Methods

Standard Fine-Tuning

Update all parametersPros:
  • Maximum accuracy
Cons:
  • Needs more memory
Best for: Production systems with good GPUs

LoRA

Low-Rank AdaptationPros:
  • 90% less memory
  • Swap adapters
  • Faster training
Best for: Large models, limited resources

QLoRA

Quantized LoRAPros:
  • Works on consumer GPUs
  • 4-bit quantization
Cons:
  • Slightly lower accuracy
Best for: Experimentation, very limited resources

Prompt/Prefix Tuning

Train only promptsPros:
  • Minimal memory
  • Very fast
Cons:
  • Limited capacity
Best for: Few-shot learning, multiple tasks

​
Progressive Approach

1

Stage 1: Prompt Engineering

Start simple, test fast
  • Test the concept
  • Gather user feedback
  • Identify limitations
  • Collect training data
2

Stage 2: Few-Shot Fine-Tuning

Improve with examples
  • Use collected examples
  • Improve consistency
  • Reduce prompt complexity
  • Validate approach
3

Stage 3: Full Fine-Tuning

Scale for production
  • Scale with more data
  • Optimize performance
  • Reduce inference costs
  • Production deployment
4

Stage 4: Continuous Improvement

Keep getting better
  • Collect production data
  • Periodic retraining
  • A/B testing
  • Performance monitoring

​
Cost Considerations

Training Cost: $0Inference: $0.01-0.10 per 1K tokensBest for: Low volume, experimentationMonthly estimate (1M tokens): $10-100
Training Cost: $10-1,000 (one-time)Inference: $0.001-0.01 per 1K tokensBest for: High volume, productionMonthly estimate (1M tokens): $1-10 + hosting
Training Cost: $10,000-millionsInference: Variable based on sizeBest for: Foundation model creatorsNot recommended unless you’re OpenAI/Google

​
Data Requirements

​
Prompt Engineering

  • Minimum: Zero-shot (no examples)
  • Better: Few-shot (3-5 examples)
  • Best: Many-shot (10+ examples in context)

​
Fine-Tuning

  • Minimum: 50-100 examples
  • Better: 500-1,000 examples
  • Best: 5,000+ examples

​
Training from Scratch

  • Minimum: 1M+ examples
  • Better: 100M+ examples
  • Best: Billions of examples

​
Quality vs Quantity Trade-offs

​
High Quality, Low Quantity

→ Fine-tune with careful data curation
  • Hand-picked examples
  • Expert annotations
  • Data augmentation

​
Low Quality, High Quantity

→ Use larger models with filtering
  • Automated cleaning
  • Statistical filtering
  • Ensemble methods

​
Mixed Quality

→ Progressive filtering approach
  • Start with all data
  • Identify quality indicators
  • Weight by quality

​
Common Mistakes to Avoid

Jumping to Training Too FastAlways try prompt engineering first. You might not need training at all.
Insufficient Training DataFine-tuning with 10 examples won’t work. Need at least 100 quality examples.
Over-Engineering SolutionsDon’t train from scratch for standard tasks. Use pre-trained models.
Ignoring MaintenanceModels need updates. Plan for retraining from day one.

​
Hybrid Approaches

​
RAG (Retrieval Augmented Generation)

Combine prompting with external knowledge. Use when:
  • Need updatable knowledge
  • Can’t fine-tune frequently
  • Have structured data

​
Ensemble Methods

Combine multiple approaches. Example:
  • Prompt for creativity
  • Fine-tuned model for accuracy
  • Vote/combine outputs

​
Chain of Thought + Fine-Tuning

Fine-tune on reasoning steps. Use when:
  • Need explainable outputs
  • Complex reasoning tasks
  • Educational applications

​
Making the Decision

​
Questions to Ask

  1. What’s my budget?
    • Low → Prompt engineering
    • Medium → Fine-tuning
    • High → Consider all options
  2. How much data do I have?
    • Less than 100 examples → Prompt engineering
    • 100-10K → Fine-tuning
    • More than 1M → Could train from scratch
  3. How unique is my task?
    • Common → Prompt engineering
    • Specialized → Fine-tuning
    • Novel → Training from scratch
  4. What accuracy do I need?
    • Acceptable → Prompt engineering
    • High → Fine-tuning
    • Perfect → Multiple iterations
  5. How fast do I need results?
    • Today → Prompt engineering
    • This week → Fine-tuning
    • This quarter → Any approach

​
Next Steps

Ready to start training?

Quick Start

Try your first model

Choose Interface

Pick UI, CLI, or API

Model Types

Select architecture