# Continuous customization
<a name="customizing-models-continuous"></a>

With **continuous customization**, you can build on a previously customized model by applying additional training — either with a *different technique* or the *same technique* with new data. For example, start with [SFT](customizing-models-sft.md) to teach domain knowledge, then apply [DPO](customizing-models-dpo.md) to align with user preferences, then apply [RLVR](customizing-models-rft.md) to improve factual accuracy.

**Important**  
Continuous customization is different from [MTRL](model-customize-mtrl.md) (Multi-Turn Reinforcement Learning). Continuous customization chains training jobs across techniques. MTRL trains agents for multi-step tool-use tasks within a single training job.

## How it works
<a name="continuous-how-it-works"></a>
+ Loads [LoRA](customizing-models-lora.md) adapter weights from a previous training job
+ Resets optimizer, scheduler, and step counter (fresh training with new technique or data)
+ The previous job's output becomes the starting point for the new training run

**Note**  
Continuous customization is currently available through the [MTRL](model-customize-mtrl.md) and continuous customization workflows using the Python SDK only. It is supported with the [LoRA](customizing-models-lora.md) training type. Continuous customization across different techniques with [FFT](customizing-models-fft.md) is not supported.

## Supported technique transitions
<a name="continuous-supported-transitions"></a>

The following table shows all supported combinations for continuous customization:


| From technique | To technique | Supported | 
| --- | --- | --- | 
| SFT | DPO | ✓ | 
| SFT | RLVR | ✓ | 
| SFT | RLAIF | ✓ | 
| SFT | SFT (new data) | ✓ | 
| SFT | MTRL | ✓ | 
| DPO | DPO (new data) | ✓ | 
| DPO | RLVR | ✓ | 
| DPO | RLAIF | ✓ | 
| DPO | SFT | ✓ | 
| RLVR | DPO | ✓ | 
| RLVR | RLVR (new data) | ✓ | 
| RLAIF | DPO | ✓ | 
| RLAIF | RLAIF (new data) | ✓ | 

## Continuous customization vs resume training
<a name="continuous-vs-resume"></a>


|  | Continuous customization | Resume training | 
| --- | --- | --- | 
| Technique | Different or same technique | Same technique only | 
| Optimizer state | Reset (fresh start) | Restored from checkpoint | 
| Use case | Chain techniques (SFT → DPO → RLVR) | Continue interrupted training | 
| Training type | [LoRA](customizing-models-lora.md) only | [LoRA](customizing-models-lora.md) only | 

## Getting started
<a name="continuous-getting-started"></a>

Continuous customization is available through the Python SDK. Provide the output path of a previous training job as the `resume_from_path` for the next technique.

```
from sagemaker.modules.train import DPOTrainer

# Continue from a previous SFT job with DPO
trainer = DPOTrainer(
    model_id="meta-textgeneration-llama-3-1-8b-instruct",
    train_data="s3://my-bucket/dpo-preferences.jsonl",
    resume_from_path="s3://my-bucket/previous-sft-output/",
    hyperparameters={
        "max_epochs": 2,
        "learning_rate": 1e-5,
        "lora_rank": 16,
    }
)
trainer.train()
```

## Example workflows
<a name="continuous-example-workflows"></a>

SFT → DPO  
Teach domain knowledge first, then align with user preferences

SFT → RLVR  
Teach task format first, then optimize for factual correctness

SFT → DPO → RLVR  
Full pipeline — knowledge → preferences → accuracy

SFT → MTRL  
Teach base capabilities, then train for multi-turn agentic tasks

DPO → DPO (new data)  
Iteratively refine preferences with new feedback data