[Rust Frontend][gRPC] Advertise LoRA capabilities (#52031)
Signed-off-by: Connor Carpenter <connorc@nvidia.com> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
This commit is contained in:
co-authored by
mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
parent
d1e3eee6fb
commit
f08a95f8d8
@@ -88,6 +88,8 @@ class EngineCoreReadyResponse:
|
||||
max_num_seqs: int
|
||||
max_num_batched_tokens: int
|
||||
instance_id: str
|
||||
supports_lora: bool
|
||||
max_loras: int
|
||||
# KV cache capacity (None for encoder-only/attention-free models).
|
||||
kv_cache_size_tokens: int | None = None
|
||||
kv_cache_max_concurrency: float | None = None
|
||||
|
||||
@@ -1636,6 +1636,12 @@ class EngineCoreProc(EngineCore):
|
||||
max_num_seqs=scheduler_config.max_num_seqs,
|
||||
max_num_batched_tokens=scheduler_config.max_num_batched_tokens,
|
||||
instance_id=self.vllm_config.instance_id,
|
||||
supports_lora=self.vllm_config.lora_config is not None,
|
||||
max_loras=(
|
||||
self.vllm_config.lora_config.max_loras
|
||||
if self.vllm_config.lora_config is not None
|
||||
else 0
|
||||
),
|
||||
kv_events_config=self.scheduler.get_kv_event_publisher_config(),
|
||||
weight_transfer_backend=(
|
||||
self.vllm_config.weight_transfer_config.backend
|
||||
|
||||
Reference in New Issue
Block a user