Python class
KVConnectorConfig
KVConnectorConfig
class max.pipelines.modeling.kv_cache_config.KVConnectorConfig(*, config_file=None, section_name=None, host_kvcache_swap_space_gb=50.0, disk_offload_dir=None, use_debug_tiered_mode=False, disk_offload_max_gb=50.0, disk_offload_direct_io=False, block_store_endpoint=None, **extra_data)
Bases: ConfigFileModel
Connector-specific configuration for KV cache connectors.
Common fields are typed. Additional connector-specific fields pass
through via extra="allow" and are accessible via model_extra.
-
Parameters:
block_store_endpoint
Endpoint for the co-located dKV service.
Remote dKV endpoints are discovered at runtime through the
Orchestrator (via external_block_metadata on the request
context), not configured statically. For multi-store reads, the
discovered metadata must include MAX-native transfer-engine metadata so
the connector can reuse KVTransferEngine.connect().
config_file
Path to the configuration file.
disk_offload_dir
Directory for disk-based KV cache offloading.
disk_offload_direct_io
disk_offload_direct_io: bool
Whether to use O_DIRECT for disk I/O.
disk_offload_max_gb
disk_offload_max_gb: float
Maximum disk space in GB for KV cache offloading.
host_kvcache_swap_space_gb
host_kvcache_swap_space_gb: float
Host memory in GiB for KV cache swapping.
model_config
model_config: ClassVar[ConfigDict] = {'extra': 'allow', 'strict': False}
Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].
section_name
Optional section name for comprehensive/multi-section config files.
If not provided, values are loaded from the YAML top-level (treating the file as an “individual config” file).
use_debug_tiered_mode
use_debug_tiered_mode: bool
Whether to use the debug tiered mode.
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!