fal launches Veo 3.1 and cuts Veo-class video pricing ~50-60%
fal added Google's Veo 3.1 (4K, native audio, video extension) and repriced the whole Veo line: Standard drops to $0.20/s no-audio ($0.40/s with audio) from $0.50/$0.75, Fast to $0.10/s ($0.15/s) from $0.25/$0.40. The old Veo 3 and Veo 3 Fast endpoints are now deprecated.
Veo 3 (fal-ai/veo3): $0.50/s audio off, $0.75/s audio on. Veo 3 Fast (fal-ai/veo3/fast): $0.25/s audio off, $0.40/s audio on. Only 1080p, 8s clips, no image-to-video or extend-video modes.
Veo 3.1 (fal-ai/veo3.1) Standard: $0.20/s audio off, $0.40/s audio on at 720p/1080p ($0.40/$0.60 at 4K). Veo 3.1 Fast: $0.10/s audio off, $0.15/s audio on at 720p/1080p ($0.30/$0.35 at 4K). Adds true 4K output, image-to-video, first/last-frame, reference-to-video, and extend-video (up to ~148s). The old fal-ai/veo3 and fal-ai/veo3/fast endpoints now show 'This endpoint is deprecated — This model is no longer supported' and display the same lower Veo 3.1-equivalent rates rather than their original $0.50/$0.75 and $0.25/$0.40 prices.
fal added Google DeepMind’s Veo 3.1 across nine endpoints (text-to-video, image-to-video, first/last-frame, reference-to-video, and extend-video, each in Standard and Fast tiers) and retired the original Veo 3 and Veo 3 Fast endpoints in the same move. The old endpoints still resolve, but each now carries a “This endpoint is deprecated — This model is no longer supported” banner and quotes the new, lower Veo 3.1-equivalent per-second rate rather than the price its own marketing copy still describes.
The repricing is steep: Standard-tier video without audio falls from $0.50/s to $0.20/s (a 60% cut), and with audio from $0.75/s to $0.40/s (47%) at 720p/1080p — fal’s new 4K tier costs $0.40/s and $0.60/s respectively, roughly what the old 1080p-only Standard rate used to be. Fast-tier drops similarly, from $0.25/$0.40 per second to $0.10/$0.15 at 720p/1080p. Veo 3.1 also adds capabilities the old endpoint never had: true 4K output, image-to-video and reference-to-video modes, and an extend-video endpoint that chains clips up to roughly 148 seconds from a single 8-second starting generation.
Notably, fal’s top-level /pricing summary table was not touched by this change — it still shows a single “Veo 3” row at $0.4/second, unchanged since 2025-07. A buyer who never clicks into the model page would not see either the new Veo 3.1 launch or the underlying price cut; the detail lives entirely on fal’s per-model pages.
The Serverless & Compute table dropped its per-second column and now publishes two hourly rates per GPU. The rates fal used to advertise as the price became the floor (H100 $1.89/h, H200 $2.10/h) while new list prices sit roughly 2x higher (H100 $3.99/h, H200 $4.50/h), so a self-serve buyer who does not contact sales sees a materially higher number. B200 lost its 'contact us' gate at $6.25/h list ($3.49/h as low as); B300 288GB ($8.50/h, $4.49/h) and RTX PRO 6000 96GB ($2.99/h, $1.10/h) joined the fleet; A100 40GB was retired. Model API rates were unchanged.