For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).
Python module
max.pipelines.lora
LoRA adapter management for MAX pipelines.
Adapter management
LoRAConfig | Configuration for LoRA (Low-Rank Adaptation) inference. |
|---|---|
LoRAManagerV3 | Loads and routes ModuleV3 LoRA adapters (adapters-as-inputs). |
LoRAModel | Manages LoRA weights and configuration for a single adapter. |
LoRATargetModule | One wrappable projection module and the PEFT projections it fuses. |
get_lora_manager | Returns the LoRA manager from the pipeline if LoRA is enabled. |
|---|
LoRA types
LoRAOperation | Enum for different LoRA operations. |
|---|---|
LoRARequest | Container for LoRA adapter requests. |
LoRAResponse | Response from LoRA operations. |
LoRAStatus | Enum for LoRA operation status. |
LoRAType | Enumeration for LoRA Types. |
ADAPTER_CONFIG_FILE | File name of the LoRA adapter's config file inside the adapter's local directory, read when the adapter is loaded. |
|---|---|
LORA_REQUEST_ENDPOINT | Name of the ZeroMQ endpoint the API server uses to push LoRA load and unload requests to workers. |
LORA_RESPONSE_ENDPOINT | Name of the ZeroMQ endpoint workers use to return LoRA operation results to the API server. |
NUM_ROUTING_INPUTS | Number of routing tensors in ModuleV3 LoRA batch info, matching max.nn.LoRA.set_lora_batch_info(). |