01
References at generation time
H3 Omni Ref lets one MiniMax H3 generation take text, images, video and audio together as references. Nuva Lab is the exclusive launch partner, and it is in private early access.
Blog · October 8, 2026
Fine-tune MiniMax H3 on your brand: LoRA, post-training and references
A general video model does not know your characters, your products or your house style. Customization closes that gap. Nuva Lab adapts open-weight models such as MiniMax H3 to your visual language. The goal is that more clips pass your review on the first try.
Why customize
Where general models fail in production
| Workload | What must stay consistent |
|---|---|
| Performance ads | Products, logos, packaging and brand colors across many variants |
| Game marketing | Characters, art style and gameplay look across creative tests |
| Short drama | Characters, wardrobe and setting across shots and episodes |
Each rejected clip costs a full generation and a review. Consistency problems are the most common reason for a reject, so they drive cost.
How we customize
Three layers of customization
02
Post-training on your data
We fine-tune or post-train the model on your approved clips and assets, so your look becomes the model's default instead of a long prompt.
03
Evaluation on your criteria
We build a test set from your own review decisions and check every model update against it before it reaches production.
Choose the method
LoRA training vs post-training vs references
| Method | What changes | Best for |
|---|---|---|
| LoRA adapter | A small set of added weights; the base model stays the same | One character, product or style |
| Full post-training | The model weights | A house style across many workloads |
| H3 Omni Ref references | No weights; references go in at generation time | Per-job consistency and fast changes |
Most teams start with references and LoRA adapters, then move to post-training when one style covers many workloads. We choose the method from your review data, not from a fixed plan.
Ownership
Your data, your model, your deployment
Your assets, your customized weights and your review data stay in your dedicated deployment. The customized model runs on a host that only you use.
Customization follows the base model license. For MiniMax H3, fine-tunes and LoRAs are Model Derivatives, so the H3 territory rule applies. Nuva Lab holds a MiniMax authorization for H3 deployments that serve US companies.
FAQ
Questions
Can I fine-tune MiniMax H3 on my own data?
Yes. H3 is open weight, so it can be fine-tuned or post-trained. The result is a Model Derivative under the MiniMax H3 license, so the license territory rule and restrictions still apply.
How do you train a LoRA for MiniMax H3?
Collect approved clips and images of one subject or style, caption them, train a LoRA adapter on the open H3 weights, and test it on held-out briefs. Nuva Lab runs this process on your dedicated deployment.
Is training a LoRA expensive?
A LoRA trains only a small set of added weights, so it needs much less compute than full post-training. Data preparation and evaluation are often the larger cost. The total depends on clip count, resolution and length.
What is H3 Omni Ref?
H3 Omni Ref lets one MiniMax H3 generation take text, images, video and audio together as references. Nuva Lab is the exclusive launch partner. It is in private early access.
How do you measure whether customization works?
We measure the acceptance rate on a held-out set of your own briefs, scored with your review criteria, before and after each change.
Sources and further reading
Read next
Make the model yours
Send us a sample of your briefs and approved work. We'll follow up about customization and H3 Omni Ref early access.
Get started · 1 min