ef6abf3dc7
Adds ALL files needed for Dockerfile build:
- qwen3_6_scripts/ (baseline patches + our optimizations)
- vllm/ (full vllm package)
- paged_attention_v2_pytorch.py (V2 with single-bmm optimization)
- Dockerfile + computility-run.yaml
Our optimizations vs baseline:
1. paged_attn.py: pre-gathered context KV (eliminates 194 gather calls),
Triton try/fallback, V2 heuristic, threshold 32K→64K
2. paged_attention_v2_pytorch.py: fills NotImplementedError,
single-bmm Phase 1 (195 launches → 3)
3. patch_enable_triton.py: HAS_TRITON=True with safety fallback
4. patch_triton_tuning.py: BLOCK=64, NUM_WARPS=4 for BI-V100
5. computility-run.yaml: gpu-memory-utilization 0.9→0.95,
max-num-batched-tokens 8192→16384
This repo can now be submitted to dev.modelhub.org.cn as-is.
35 lines
799 B
Python
35 lines
799 B
Python
import msgspec
|
|
|
|
from vllm.adapter_commons.request import AdapterRequest
|
|
|
|
|
|
class PromptAdapterRequest(
|
|
msgspec.Struct,
|
|
array_like=True, # type: ignore[call-arg]
|
|
omit_defaults=True, # type: ignore[call-arg]
|
|
frozen=True): # type: ignore[call-arg]
|
|
"""
|
|
Request for a Prompt adapter.
|
|
"""
|
|
__metaclass__ = AdapterRequest
|
|
|
|
prompt_adapter_name: str
|
|
prompt_adapter_id: int
|
|
prompt_adapter_local_path: str
|
|
prompt_adapter_num_virtual_tokens: int
|
|
|
|
def __hash__(self):
|
|
return super().__hash__()
|
|
|
|
@property
|
|
def adapter_id(self):
|
|
return self.prompt_adapter_id
|
|
|
|
@property
|
|
def name(self):
|
|
return self.prompt_adapter_name
|
|
|
|
@property
|
|
def local_path(self):
|
|
return self.prompt_adapter_local_path
|