Open madroidmaq opened 3 weeks ago
Add the stream_generate_image function to expose the results of each step of the inference for follow-up GUI interaction.
stream_generate_image
The usage would look something like this:
def stream_demo(args): seed = int(time.time()) if args.seed is None else args.seed flux = Flux1.from_alias(args.model) stream = flux.stream_generate_image( seed=seed, prompt=args.prompt, report_step=5, config=Config( num_inference_steps=args.steps, height=args.height, width=args.width, guidance=args.guidance, ), ) step_count = 0 for image in stream: step_count += 1 intermediate_output_path = f"{args.output}_report_step_{step_count}.png" ImageUtil.save_image(image, "build/stream/" + intermediate_output_path) print(f"Generation complete. Total steps: {step_count}") stream_demo()
Add the
stream_generate_image
function to expose the results of each step of the inference for follow-up GUI interaction.The usage would look something like this: