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.
31 lines
875 B
Python
31 lines
875 B
Python
from dataclasses import dataclass
|
|
from typing import Literal
|
|
|
|
import torch
|
|
from PIL import Image
|
|
|
|
from vllm.assets.base import get_vllm_public_assets
|
|
|
|
VLM_IMAGES_DIR = "vision_model_images"
|
|
|
|
|
|
@dataclass(frozen=True)
|
|
class ImageAsset:
|
|
name: Literal["stop_sign", "cherry_blossom"]
|
|
|
|
@property
|
|
def pil_image(self) -> Image.Image:
|
|
|
|
image_path = get_vllm_public_assets(filename=f"{self.name}.jpg",
|
|
s3_prefix=VLM_IMAGES_DIR)
|
|
return Image.open(image_path)
|
|
|
|
@property
|
|
def image_embeds(self) -> torch.Tensor:
|
|
"""
|
|
Image embeddings, only used for testing purposes with llava 1.5.
|
|
"""
|
|
image_path = get_vllm_public_assets(filename=f"{self.name}.pt",
|
|
s3_prefix=VLM_IMAGES_DIR)
|
|
return torch.load(image_path, weights_only=True)
|