xAI Imagine, Voice and Transcription: Why Seconds, Images and Hours Cannot Be Averaged

Video output is a priced duration

The xAI Imagine rate card specifies output-video seconds by resolution and model. For grok-imagine-video, it lists $0.05 per output second at 480p and $0.07 at 720p; some source media can produce separate input charges. A ten-second 720p output therefore has a listed output component of $0.70, not a fixed ten-credit consumption. The quoted figure is not a promise that one prompt yields ten usable seconds.

Image generation is not per-second

The same card lists grok-imagine-image output at $0.02 per 1K or 2K generated image, while other Imagine quality tiers have different image rates. Input source images can carry their own line item. Store input media units separately from delivered images; changing the number of references can change price even when one final image is exported. Do not extrapolate a generative image rate to video or resize workflows.

Speech-to-text versus real-time speech

xAI lists speech-to-text REST at $0.10 per hour and streaming at $0.20 per hour, speech-to-speech at $0.08 per minute, and text-to-speech at $15 per million characters. Hours, minutes and characters are four different denominators. A streaming support call may add model processing, phone network costs and failure retries outside the nominal recognition charge. Preserve STT as a separate product, not a token conversion.

Model versions and access gates

The official release notes track a sequence of Imagine product updates. Do not interpret “starting at” prices as applicable to every current model, output quality or combination of reference media. Some models may be preview-only, retired, or have default quality settings that change with releases. The exact output resolution, model version, media modality and API region are necessary to reproduce a bill or forecast a pipeline reliably.

Budget a realistic clip pipeline

For a planned 120 seconds of finished footage, six 10-second attempts per minute of accepted footage would require much more generated material than the final output. Forecast each attempt at its own resolution and billed duration, then add image reference charges, upscaling, storage, review and audio. The accepted-footage yield is a production estimate you must measure, not a parameter that the vendor’s tariff specifies. Separate image, video, transcription and speech budgets.

How to verify the actual invoice

Begin with a small representative workload rather than the cheapest published headline. Record the exact provider and billing channel, API model identifier, currency, account geography, plan and commitment length. The price or quota evidence in this guide is keyed to generated_image, output_video_second, speech_to_speech_minute, transcribed_audio_hour, tts_characters_per_1m. Do not combine proprietary creative credits, token meters or media seconds into a single currency or invent a conversion rate between them. If an enterprise or regional rate is not published, leave its price unknown and obtain a written quote.

Keep separate estimates for failed attempts, optional tools, storage, exports, human quality review and tax. Use source-verified line items from the companion CSV instead of typing permanent numbers into your forecasting formulas. Where an old plan or model was superseded, preserve its original historical record and create an additive later snapshot. This guide does not claim a product test, performance ranking, independent legal assessment or live procurement approval.

Availability, alternatives and editorial limits

Available plans, supported endpoints, trial eligibility, quota reset and contract rights may vary by region and account even when a public website displays a single global currency. Compare an alternative only with the same accepted task, input/output volume, model version and privacy requirements. Vendor-provided marketing claims are not independent evidence of quality. This historical pricing article may be cross-referenced by the model, media or enterprise procurement team, but its purpose is to explain the billing mechanics without duplicating their evaluations or inventing test scores.