Kling 2.6 Serverless API

 1# pip install "segmind>=1.1.0"
 2# export SEGMIND_API_KEY="YOUR_API_KEY"
 3from segmind import SegmindClient, InferenceFailed, InferenceTimeout
 4
 5# Async (v2) — recommended for long-running / video models.
 6# run() blocks up to 600s; submit_async + job.wait(timeout=...) sets a longer
 7# deadline and keeps the request_id so you can re-poll later.
 8client = SegmindClient()                      # reads SEGMIND_API_KEY
 9payload = {
10    "prompt": "Visual: São Paulo, Brazil – a narrow street packed with vibrant murals in neon greens, reds, and yellows. The atmosphere is electric: street vendors, kids playing football, and local dancers vibing in the background. The city feels loud, alive, and unapologetic. Subject: A Black male rapper in raw streetwear — fitted cap, graphic tee, worn sneakers, simple chain. He stands front and center, locked in with the camera, body moving naturally to the rhythm, commanding the street like it’s his stage. Audio: [Male rapper, high-energy, gritty voice] Rapping over a hard street beat blended with Brazilian percussion: “Painted walls talk loud, yeah, they know my name, Came up from the block where the heat stay flame. City never soft, had to hustle, stay sharp, Turned struggle into fire, now I’m leavin’ my mark. Bass hit heavy, feel the ground when I step, From these streets to the world, yeah, I’m reppin’ respect.” Background: Heavy bass layered with live drum hits, claps, and subtle street noise — bikes passing, distant voices, raw urban texture. Camera: Rapid cuts between tight close-ups of his face and hands, wide shots of the colorful murals, quick flashes of dancers and street life. Handheld movement keeps it gritty and real, ending on a strong close-up as the beat hits.",
11    "negative_prompt": "no noise, no distortion, no clutter",
12    "cfg_scale": 0.8,
13    "duration": "10",
14    "mode": "pro",
15    "aspect_ratio": "16:9",
16    "generate_audio": True,
17}
18job = client.submit_async("kling-2.6", **payload)
19print(job.request_id)                         # available immediately
20try:
21    result = job.wait(timeout=900, interval=2.0)
22    print(result["status"])                  # COMPLETED
23    print(result.get("output"))              # model output (e.g. video URL)
24except InferenceTimeout as e:
25    print("still running:", e.request_id)    # re-poll later with this id
26except InferenceFailed as e:
27    print("failed:", e.detail)
28
29# Fast models (<=600s) can use the one-liner instead:
30# result = segmind.run("kling-2.6", **payload)