[ModelRunner V2] Support encoder-only attention (#49331)

Signed-off-by: Nick Hill <nickhill123@gmail.com>
Signed-off-by: Taneem Ibrahim <taneem.ibrahim@gmail.com>
Co-authored-by: Taneem Ibrahim <taneem.ibrahim@gmail.com>
This commit is contained in:
Nick Hill
2026-07-27 00:38:44 +00:00
committed by GitHub
co-authored by Taneem Ibrahim
parent 0934b26790
commit fdaa0d9e59
8 changed files with 276 additions and 4 deletions
+3
View File
@@ -459,6 +459,9 @@ class GPUModelRunner(LoRAModelRunnerMixin):
self.attn_groups, attn_cg_support, self.kernel_block_sizes = init_attn_backend(
self.kv_cache_config, self.vllm_config, self.device
)
attn_cg_support = attn_cg_support.narrow(
*self.model_state.get_additional_cg_support()
)
self.block_tables = BlockTables(
block_sizes=block_sizes,
max_num_reqs=self.max_num_reqs,