Ivor StrelarandClaude
608f52915c
- Add RunwayML Upscale V1 node for enhanced video upscaling - Add WAN 2.2 I2V 720p node for high-quality image-to-video generation - Both nodes follow established WaveSpeed patterns: - Use real_client.post() directly instead of request objects - Include sync/async mode support for user control - Comprehensive error handling and progress logging - Proper input validation with forceInput for URLs - Consistent return types and naming conventions - Include corresponding request classes for future extensibility 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
94 lines
3.5 KiB
Python
94 lines
3.5 KiB
Python
import time
|
|
from .wavespeed_api.client import WaveSpeedClient
|
|
|
|
class RunwaymlUpscaleV1Node:
|
|
"""
|
|
WaveSpeed AI RunwayML Upscale V1 Node
|
|
|
|
Enhanced video upscaling using RunwayML's advanced AI models.
|
|
Improves video resolution and quality while preserving content integrity.
|
|
"""
|
|
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {
|
|
"required": {
|
|
"client": ("WAVESPEED_AI_API_CLIENT",),
|
|
"video_url": ("STRING", {
|
|
"default": "",
|
|
"tooltip": "URL of input video for upscaling (connect from Upload Video node)",
|
|
"forceInput": True
|
|
}),
|
|
"enable_sync_mode": ("BOOLEAN", {
|
|
"default": True,
|
|
"tooltip": "Wait for upscaling to complete before returning"
|
|
}),
|
|
},
|
|
}
|
|
|
|
RETURN_TYPES = ("STRING",)
|
|
RETURN_NAMES = ("upscaled_video_url",)
|
|
CATEGORY = "WaveSpeedAI"
|
|
FUNCTION = "execute"
|
|
|
|
def execute(self, client, video_url, enable_sync_mode):
|
|
# Create the actual client object from the client dict
|
|
real_client = WaveSpeedClient(api_key=client["api_key"])
|
|
|
|
# Build payload
|
|
payload = {
|
|
"video": video_url
|
|
}
|
|
|
|
# API endpoint
|
|
endpoint = "/api/v3/runwayml/upscale-v1"
|
|
|
|
try:
|
|
print(f"Submitting RunwayML Upscale V1 task...")
|
|
print(f"Video URL: {video_url[:50]}...")
|
|
|
|
response = real_client.post(endpoint, payload, timeout=real_client.once_timeout)
|
|
|
|
if enable_sync_mode:
|
|
# In sync mode, wait for the result
|
|
if "outputs" in response and response["outputs"]:
|
|
video_urls = response["outputs"] # Already a list
|
|
print(f"Task completed successfully. Received {len(video_urls)} video(s)")
|
|
# Return the first video URL
|
|
video_url = video_urls[0]
|
|
return (video_url,)
|
|
else:
|
|
raise Exception(f"No output received from sync API. Response: {response}")
|
|
else:
|
|
# In async mode, submit task and poll for results
|
|
task_id = response["id"]
|
|
print(f"Task submitted successfully. Request ID: {task_id}")
|
|
|
|
try:
|
|
# Poll for task completion - longer timeout for upscaling
|
|
result = real_client.wait_for_task(task_id, polling_interval=2, timeout=600)
|
|
|
|
if "outputs" in result and result["outputs"]:
|
|
video_urls = result["outputs"] # Already a list
|
|
print(f"Task completed successfully. Received {len(video_urls)} video(s)")
|
|
# Return the first video URL
|
|
video_url = video_urls[0]
|
|
return (video_url,)
|
|
else:
|
|
raise Exception(f"Task completed but no output received. Response: {result}")
|
|
|
|
except Exception as e:
|
|
raise Exception(f"Async task failed: {str(e)}")
|
|
|
|
except Exception as e:
|
|
print(f"Error in RunwayML Upscale V1: {str(e)}")
|
|
raise e
|
|
|
|
# Add the mappings that your __init__.py file will automatically find
|
|
NODE_CLASS_MAPPINGS = {
|
|
"WaveSpeedAI_RunwaymlUpscaleV1Node": RunwaymlUpscaleV1Node
|
|
}
|
|
|
|
NODE_DISPLAY_NAME_MAPPINGS = {
|
|
"WaveSpeedAI_RunwaymlUpscaleV1Node": "WaveSpeedAI RunwayML Upscale V1"
|
|
} |