Skip to main content

GMICloud media provider adapters for genblaze (video, image, audio)

Project description

genblaze-gmicloud

GMICloud multi-provider video / image / audio adapters for genblaze — access Seedance, Kling, Veo, Sora, Wan, Seedream, FLUX, Gemini image, ElevenLabs, MiniMax and more through one API with SHA-256 provenance manifests.

genblaze-gmicloud wraps GMICloud's request-queue API, giving you one-call access to a large catalog of video, image, and audio models — including Kling, Veo, Sora, Wan, Seedream, FLUX-Kontext-Pro, Gemini-2.5-Flash-Image, ElevenLabs TTS, MiniMax TTS, and MiniMax Music — via three genblaze provider classes. Compose into multi-step AI pipelines, persist outputs to Backblaze B2 or any S3-compatible store, and emit a tamper-evident provenance manifest on every run.

Why genblaze-gmicloud

  • One API, dozens of models — text-to-video (Seedance, Kling, Veo, Sora, Wan), text-to-image (Seedream, FLUX, Gemini, Reve), audio (ElevenLabs, MiniMax TTS/Music).
  • Provenance by default — SHA-256-verified manifest with provider, model, prompt, params, cost.
  • Cost trackingstep.cost_usd is populated from GMICloud's response.
  • Two auth modes — API key (GMI_API_KEY) or SDK email/password.
  • Production-ready — retries, timeouts, progress streaming, step caching.
  • Durable storage — plug genblaze-s3 in for Backblaze B2 / AWS S3 / R2 / MinIO persistence.

Providers + models

Provider class Modality Example models
GMICloudVideoProvider video Kling-Text2Video-V1.6-Pro, Kling-Image2Video-V2.1-Master, Veo3, Wan-2.6-T2V, Seedance, Sora
GMICloudImageProvider image Seedream-5.0-Lite, Gemini-2.5-Flash-Image, Reve-Edit-Fast, FLUX-Kontext-Pro
GMICloudAudioProvider audio ElevenLabs-TTS-v3, MiniMax-TTS-Speech-2.6-Turbo, MiniMax-Music-2.5

Registered via entry points as gmicloud, gmicloud-image, and gmicloud-audio. Any model on GMICloud's queue is supported — pass the exact model slug.

Install

pip install genblaze-gmicloud

Quickstart — video (Kling)

pip install genblaze-core genblaze-gmicloud
export GMI_API_KEY="..."
from genblaze_core import Modality, Pipeline
from genblaze_gmicloud import GMICloudVideoProvider

run, manifest = (
    Pipeline("gmicloud-video-demo")
    .step(GMICloudVideoProvider(), model="Kling-Text2Video-V1.6-Pro",
          prompt="A drone shot flying over a misty mountain valley at sunrise, cinematic",
          modality=Modality.VIDEO, duration=10, aspect_ratio="16:9")
    .run(timeout=600)
)
print(run.steps[0].assets[0].url, f"${run.steps[0].cost_usd:.3f}")

Quickstart — image (Seedream)

from genblaze_gmicloud import GMICloudImageProvider

run, manifest = (
    Pipeline("gmicloud-image-demo")
    .step(GMICloudImageProvider(), model="Seedream-5.0-Lite",
          prompt="A photorealistic macro shot of morning dew on a spider web, soft bokeh",
          modality=Modality.IMAGE, aspect_ratio="16:9")
    .run(timeout=120)
)

Quickstart — audio (ElevenLabs via GMICloud)

from genblaze_gmicloud import GMICloudAudioProvider

run, manifest = (
    Pipeline("gmicloud-audio-demo")
    .step(GMICloudAudioProvider(), model="ElevenLabs-TTS-v3",
          prompt="Welcome to Genblaze — the fastest way to build generative AI pipelines.",
          modality=Modality.AUDIO)
    .run(timeout=120)
)

Persist to Backblaze B2

from genblaze_core import KeyStrategy, ObjectStorageSink
from genblaze_s3 import S3StorageBackend

storage = ObjectStorageSink(
    S3StorageBackend.for_backblaze("my-bucket"),
    key_strategy=KeyStrategy.HIERARCHICAL,
)
# pass sink=storage to .run(…)

Backblaze B2 is the recommended default sink — cost-efficient, S3-compatible, Object Lock for immutable manifests.

Credentials

Auth mode Env var
API key (recommended) GMI_API_KEYhttps://console.gmicloud.ai/
SDK email/password GMI_CLOUD_EMAIL + GMI_CLOUD_PASSWORD

Documentation

Related packages

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

genblaze_gmicloud-0.2.0.tar.gz (12.3 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

genblaze_gmicloud-0.2.0-py3-none-any.whl (14.6 kB view details)

Uploaded Python 3

File details

Details for the file genblaze_gmicloud-0.2.0.tar.gz.

File metadata

  • Download URL: genblaze_gmicloud-0.2.0.tar.gz
  • Upload date:
  • Size: 12.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.5

File hashes

Hashes for genblaze_gmicloud-0.2.0.tar.gz
Algorithm Hash digest
SHA256 a405482d9276ee2b6d5c33c5e2594f889bbf801a38248e4d0e96a215e9fa53be
MD5 04679af44cdb8478771c57fcf758fbad
BLAKE2b-256 0a84cdff8d22d58b42858683933563113aa0e51d2c06737f259ea9f2b6b98ace

See more details on using hashes here.

File details

Details for the file genblaze_gmicloud-0.2.0-py3-none-any.whl.

File metadata

File hashes

Hashes for genblaze_gmicloud-0.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 ef0c995e8f75b3a015d4b647d6168f119315155761f18b51214c060fc86175a1
MD5 1ead788857289870d3554529a77784a3
BLAKE2b-256 6da404742b523262bbafd927d57b15fc502d1ee9d10f75c5bd1a78f544ccb322

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page