# LoRA Fine-Tune

Low-Rank Adaptation (LoRA) adapts a model to a task by training a small set of additional weights instead of retraining the whole model. **LoRA Fine-Tune** applies that method to supported models on Azion's infrastructure, as an extension of [AI Inference](/en/documentation/platform/ai-inference/). LoRA applies to some models and not to others, and each model page states which case it is.

## Supported models

The table lists every model AI Inference runs, with what the model's own page states about LoRA. Yes and No repeat that statement. A dash means the page states no value, which is not the same as a No.

| Model                                                                                       | Supports LoRA |
| ------------------------------------------------------------------------------------------- | ------------- |
| [Mistral 3 Small (24B AWQ)](/en/documentation/platform/ai-inference/mistral-3-small/)       | No            |
| [BAAI/bge-reranker-v2-m3](/en/documentation/platform/ai-inference/baai-bge-reranker-v2-m3/) | No            |
| [InternVL3](/en/documentation/platform/ai-inference/internvl3/)                             | No            |
| [Qwen2.5 VL AWQ 3B](/en/documentation/platform/ai-inference/qwen-2-5-vl-3b/)                | Yes           |
| [Qwen2.5 VL AWQ 7B](/en/documentation/platform/ai-inference/qwen-2-5-vl-7b/)                | Yes           |
| [Qwen3 30B A3B Instruct 2507 FP8](/en/documentation/platform/ai-inference/qwen3-30ba3b/)    | Yes           |
| [Qwen3 Embedding 4B](/en/documentation/platform/ai-inference/qwen3-embedding-4b/)           | —             |
| [Nanonets-OCR-s](/en/documentation/platform/ai-inference/nanonets-ocr-s/)                   | —             |
| [GPT-OSS 20B](/en/documentation/platform/ai-inference/gpt-oss-20b/)                         | No            |

For the category, the id, the context length, and the input types of a model in this table, refer to [AI models](/en/documentation/platform/ai-inference/models/).

---

## Relationship to AI Inference

LoRA Fine-Tune is an Azion product of its own, and the product it extends is AI Inference. AI Inference supports LoRA through an add-on, and the adaptation runs on Azion's infrastructure, with no infrastructure for you to provision or manage. You are responsible for the training data, the model inputs and outputs, and the third-party license of the model you adapt.

---

## Billing metrics

LoRA Fine-Tune is billed on two metrics, Compute Time and Requests, and it is metered separately from AI Inference. Compute Time is the duration of active execution in hours multiplied by the memory allocated in gigabytes, stated in GB-hour. For the rate charged for each metric, refer to [Pricing](/en/documentation/fundamentals/pricing/#lora-fine-tune), which carries a section of its own for LoRA Fine-Tune.

---

## Related resources

- [AI models](/en/documentation/platform/ai-inference/models.md): The category, id, context length, and input types of every model AI Inference runs.
- [AI Inference](/en/documentation/platform/ai-inference.md): The product LoRA Fine-Tune extends, and the one that runs the models.
- [How AI Inference works](/en/documentation/platform/ai-inference/how-it-works.md): What executes a model, and where Azion's part of a model call ends.
- [Model invocation](/en/documentation/platform/ai-inference/model-invocation.md): The request body and the binding that a call to a model travels through.
- [Pricing](/en/documentation/fundamentals/pricing.md#lora-fine-tune): The rate charged for the Compute Time and the Requests of LoRA Fine-Tune.
- [Support](/en/documentation/support.md): Where to ask what this page does not state about adapting a model.