Add Docker run files (Dockerfile, compose, run script)

Dockerfile

@@ -0,0 +1,13 @@
+# Darwin-36B-Opus — VKAE Accelerated
+# The model weights + VKAE-accelerated serving runtime are ALREADY bundled in the
+# prebuilt image below. You normally do not need to build anything — just pull & run:
+#
+#     docker pull vidraft/darwin36-vkae:281
+#     docker run --gpus all -p 8000:8000 vidraft/darwin36-vkae:281
+#
+# This Dockerfile is only an extension point (add your own layers on top).
+# The VKAE acceleration recipe is proprietary and is not distributed in source form.
+FROM vidraft/darwin36-vkae:281
+
+# OpenAI-compatible API on port 8000 (entrypoint is baked into the base image).
+EXPOSE 8000

README.md

@@ -33,6 +33,9 @@ docker run --gpus all -p 8000:8000 vidraft/darwin36-vkae:281
 
 The container serves an **OpenAI-compatible API** on port 8000 — point any OpenAI client at `http://localhost:8000/v1`. A Blackwell (B200) or Hopper (H100/H200) class GPU is recommended.
 
+
+> 📦 **Ready-to-use files in this repo:** [`Dockerfile`](./Dockerfile), [`docker-compose.yml`](./docker-compose.yml), [`run_docker.sh`](./run_docker.sh) — pull-and-run, no build required.
+
 ## Links
 
 - **Live acceleration leaderboard** → [VIDraft/vkae](https://huggingface.co/spaces/VIDraft/vkae)

docker-compose.yml

@@ -0,0 +1,16 @@
+# Darwin-36B-Opus — VKAE Accelerated
+# Launch:  docker compose up
+services:
+  vkae:
+    image: vidraft/darwin36-vkae:281
+    ports:
+      - "8000:8000"          # OpenAI-compatible API -> http://localhost:8000/v1
+    ipc: host
+    restart: unless-stopped
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              count: all
+              capabilities: [gpu]

run_docker.sh

@@ -0,0 +1,17 @@
+#!/usr/bin/env bash
+# Darwin-36B-Opus — VKAE Accelerated : one-command launch.
+# Requires an NVIDIA GPU (Blackwell B200 / Hopper H100/H200, ~80GB) + nvidia-container-toolkit.
+set -euo pipefail
+
+IMAGE="vidraft/darwin36-vkae:281"
+
+docker pull "$IMAGE"
+docker run --rm --gpus all -p 8000:8000 --ipc=host "$IMAGE"
+
+# The container serves an OpenAI-compatible API on http://localhost:8000/v1
+#
+# Discover the served model name:
+#   curl http://localhost:8000/v1/models
+#
+# Chat completion:
+#   curl http://localhost:8000/v1/chat/completions #     -H "Content-Type: application/json" #     -d '{"model":"darwin36-vkae","messages":[{"role":"user","content":"Hello!"}]}'