Guides
Sume guides: the API, Formats, agents, models, media tools and pricing, each answer dated and sourced.
AI Act closed-loop film previs: where preview ends, output begins
The EU FAQ puts closed-loop film production outside marking scope unless the output is final. Avatar preview stills are never captioned; the final MP4 can be.
EU AI Content Code: Section 1 providers or Section 2 deployers?
Section 1 of the EU Code is for providers marking AI output; Section 2 is for deployers labelling deepfakes. A team shipping API video starts with its own role.
Making AI videos for clients: are you an AI Act deployer?
The Commission FAQ treats freelance or regular paid use of an AI system as professional, making you a deployer. Personal use is excluded. Not legal advice.
Nova Reel start_async_invoke S3 output vs a Sume result URL
Nova Reel's start_async_invoke writes output.mp4 to your S3 bucket and needs IAM. A Sume job returns a result_url and durable media.sume.com links.
Amazon Responsive eCommerce Creative: assets to supply
Amazon's REC builds ads from lifestyle images, video, logos and copy. Prepare lifestyle stills with reference-guided edits and short clips.
Android Automotive Play screenshots: 800x1280 and 1024x768
Play wants 2 portrait (800x1280) and 2 landscape (1024x768) Automotive screenshots of generic system UI. Both fit the ChatGPT Image 2.5 custom-size limits.
Android XR Play screenshots: 8:5 at 3200x2000, not 3840x2400
Play recommends 3840x2400 for Android XR, but that is 9,216,000 pixels, over the ChatGPT Image 2.5 custom-size cap. Use 3200x2000: 8:5 and under it.
Arcads clone static ad vs Sume Avatar Video product image
Arcads /clone-static-ad recreates a static ad as video. On Sume, the closest path is Avatar Video with an optional product_image and a scene prompt.
Arcads MCP out of credits vs Sume 402 insufficient_credits
Arcads says generation fails without credits. Sume answers 402 insufficient_credits before provider work, and an MCP agent can preview cost with dry_run first.
California SB 1000: no user threshold, new verification tool
SB 1000 recasts the California AI Transparency Act: no user threshold, a disclosure verification tool, no manifest option. What Sume's docs list.
SB 1050 takedown orders: what an advertising medium must do
SB 1050 bars an advertising medium from running an ad after a served court order. How ad files and URLs fit in, and what a corrected cut means.
SB 1050 clear and conspicuous disclosure: the definition
SB 1050 defines clear and conspicuous as difficult to miss and easy to read in the ad's medium. Map those words to caption contrast and placement controls.
California SB 1050: does an AI avatar count as synthetic?
SB 1050 requires a clear and conspicuous disclosure for ads that prominently include a synthetic performer. Read the definition next to avatar creation.
Captions AI avatar looks vs one Sume avatar handle per look
Captions added Avatar Looks to save new looks per avatar. On Sume you create one avatar handle per look from a prompt, a profile or a reference image.
Claude API compaction block: keep Sume job ids past the summary
The compact-2026-09-04 beta swaps old messages for a signed compaction block. Keep Sume job ids and keys outside it; re-read with jobs_status.
Claude Code MCP sign-in link stopped working: use the newest one
A repeat MCP sign-in request replaces the pending link in Claude Code, so the older link can stop working. Finish the newest Sume sign-in link only.
Workers AI paid models 20/min, 50 with prepaid vs Sume plans
Cloudflare lifts paid Workers AI models from 20 to 50 requests per minute with prepaid credits. On Sume, top-ups do not raise concurrency; the plan does.
Codex default_tools_approval_mode writes with Sume tools
With default_tools_approval_mode = "writes", Codex prompts for tools not marked read-only. Sume marks reads readOnlyHint true and writes false.
Coupang main image: white background, square, 95% fill
Coupang's main image should be square on white with the item near 95% of the frame. Request aspect_ratio 1:1 from Sume and check the fill yourself.
Devin Desktop MCP authorization opens the provider: Sume page
Devin Desktop 3.10.23 opens MCP authorization at the provider directly. For Sume that is mcp.sume.com/oauth/authorize, then a consent page on the MCP host.
Devin MCP write scopes by default: Sume keeps write opt-in
Devin now requests read/write scopes by default for Microsoft 365 MCP. Sume's hosted MCP asks for read only unless the user turns Write on at consent.
How long must the EU AI label stay on screen? Cue timing
The Commission gives no seconds: a limited-time disclosure should stay visible long enough to be read by people with cognitive or processing difficulties.
fal content_policy_violation 422 vs Sume generation_rejected
fal returns content_policy_violation as a non-retryable 422. Sume's closest job error category is generation_rejected: inspect events and fix the input.
fal generation_timeout 504 error: what to do on Sume
fal returns generation_timeout as a 504 typed error. Sume has a generation_timeout category; the documented action is to poll status or retry later.
fal queue status IN_QUEUE IN_PROGRESS COMPLETED on Sume
Sume's status endpoint returns a queue-shaped status field with IN_QUEUE, IN_PROGRESS, COMPLETED, FAILED and CANCELED, mapped one-to-one onto sume_status.
x-fal-needs-retry header: what Sume uses to signal a retry
fal marks retryable errors with X-Fal-Needs-Retry. Sume signals retry through error category and code, retry-after on 429, and the same idempotency key.
x-fal-no-retry header: safe retries with a Sume Idempotency-Key
fal retries queued requests; X-Fal-No-Retry turns that off. On Sume you keep retries and make them safe with an Idempotency-Key on every paid submit.
FFmpeg 8.1 AV1 and ProRes encoding: Sume has no codec field
FFmpeg 8.1 adds D3D12 H.264/AV1 and Vulkan ProRes encoding. Sume compiles ffmpeg server-side and rejects codec and crf fields, so you cannot pick an encoder.
FFmpeg 8.1 drawvg and vpp_amf: not on Sume's allowlist
FFmpeg 8.1 lists new drawvg and vpp_amf filters. Sume video-filter only runs allowlisted filters and refuses unknown names with unknown_filter.
Final Cut Pro convert closed captions: burn cues in Sume
Final Cut Pro 12.3 converts closed captions to subtitles. To burn existing text onto a video elsewhere, send cues or segments to Sume; SRT is unsupported.
Final Cut Pro Edit Detection: sample stills to see the cuts
Final Cut Pro 12.3 Edit Detection splits a rendered video at shot changes. Sume does not detect cuts; video-frames samples stills at an fps so you can see them.
Final Cut Pro Generate Captions: US English only, Korean fix
Apple's Generate Captions in Final Cut Pro 12.3 is U.S. English only. For Korean speech, Sume video captions takes a language hint and Hangul caption styles.
Firefly Composite API for product photos vs Sume reference edit
Firefly Composite Operations blend a product photo into a generated scene. On Sume, send the photo as an input reference, with an optional mask_url.
Firefly Custom Models API vs Sume: references, no training
Firefly Custom Models trains subject or style models you call by asset ID. Sume offers no image model training; use up to 16 input references.
Adobe Firefly Image 5 API: native 4 MP vs Sume resolution tiers
Firefly's Image5 model is described as native 4 MP with Instruct Edit. Sume does not list a Firefly model; it uses 512, 1K, 2K and 4K tiers.
Firefly Upscale API vs Sume's image_upscale_create tool
Firefly's Upscale API enhances resolution at 2x, 4x or 6x. Sume exposes upscaling as the image_upscale_create MCP tool: a paid job you poll.
drama-3-preview model id vs Sume TTS Router model ids
Fish selects drama-3-preview with a model header. Sume TTS Router takes a required model id from its catalog, and unknown ids fail with 400 model_not_found.
Fish Audio Drama 3 single-word fix vs Sume sentence segments
Fish Audio says Drama 3 preview can fix a single word. Sume has no word-level repair: regenerate a sentence and join takes with Timeline audio concat.
Flux TTS expressivity -2 to 2 vs Sume's emotion guide
Deepgram's Flux TTS expressivity runs -2 to 2 (0 nominal). Sume TTS has no such dial: generation_config takes volume, speed and a free-text emotion guide.
FTC: actors, dramatizations and scripted AI avatar ads
The FTC Q&A says actors in an obviously fictional dramatization are not giving testimonials, yet could still be deceptive. What that means for avatar ads.
Is #ad enough in a video? What the FTC Q&A says
The FTC says #Ad may work at the start of a text post but may be too easy to miss in a video. Text in a video must stand out. Sume captions burn it in.
FTC reviews rule and AI stock avatars: fake or false testimonials
The FTC Q&A says no blanket ban on AI avatars; a testimonial is barred only if fake or false. Here is how a Sume avatar and script map to that answer.
FTC unavoidable disclosure: a link or hover is not enough
Under the FTC's testimonials rule Q&A, a disclosure is avoidable if viewers must click or hover to see it. Here is how burned-in text passes that test.
Gemini 3.1 flash image preview shutdown: Sume model ids
Google lists gemini-3.1-flash-image-preview and gemini-3-pro-image-preview for June 25, 2026 shutdown. Sume callers use catalog ids like google/nano-banana-2.
Gemini static vs dynamic webhooks: Sume sets one URL per job
Gemini splits project-level static webhooks from per-request dynamic ones. Sume has only the per-job kind: pass webhook_url on each submit call.
Gemini JWKS JWT webhook signatures vs Sume HMAC SHA-256
Gemini dynamic webhooks use asymmetric JWKS-signed JWTs. Sume signs the raw body with a shared-secret HMAC SHA-256 and a timestamp. How each verifies.
Gemini last_event_id stream resume vs Sume: no SSE, poll status_url
Gemini background interactions can resume a dropped stream from the last event. Sume has no stream to resume: poll status_url and read events_url snapshots.
Gemini Omni videos over 4MB: poll to ACTIVE vs a Sume result URL
Gemini Omni returns large videos as a URI you poll until ACTIVE. Sume results are durable media.sume.com URLs with no ACTIVE state and no expiry to manage.