tmp_chute.py

import os
from chutes.chute import NodeSelector
from chutes.chute.template.vllm import build_vllm_chute
os.environ["NO_PROXY"] = "localhost,127.0.0.1"

chute = build_vllm_chute(
    username="achoji",
    readme="ronx-labs/affine-081410",
    model_name="ronx-labs/affine-081410",
    image="chutes/vllm_gptoss:0.10.1.dev5",
    concurrency=16,
    revision="606d2f4e4f62a2d549de4daa8c602930dcae3e51",
    node_selector=NodeSelector(
        gpu_count=8,
        min_vram_gb_per_gpu=24,
    ),
)