Black Forest LabsBlack Forest Labs·🎬 Video Generation·New

Flux 3 First Last Frame

anonymized
Try on Venice.ai ↗
Quick reference
Flux 3 First Last Frame — TLDR
  • 🆕 Keyframe video mode of FLUX 3, Black Forest Labs' multimodal image-video-audio model
  • 🎯 Supply a first and last frame; model generates the transition
  • 💬 FLUX 3 video output arrives with natively synchronized audio
  • 🔧 Clips can be chained into longer multi-shot sequences
  • 🏢 Earlier FLUX 2 entries here cover still-image generation and editing only
  • 🧠 Vendor reports gains over earlier FLUX versions on complex prompts, text
  • 🌐 Sibling endpoints cover text-to-video and single-image-to-video
  • 📏 Resolution, frame rate and maximum clip length not specified in catalog
💰 Pricing
$0.940 – $6.38
per generation
📅 On Venice since
Aug 4, 2026
0 days ago
Provider

Black Forest Labs is a generative AI company based in Freiburg im Breisgau, Germany, founded by former members of Stability AI. The lab is best known for developing the Flux family of text-to-image models, which generate images from natural language prompts…

Read full profile →
6 models on Venice
3 video · 2 image · 1 inpaint
Since Nov 25, 2025

About this model

Flux 3 First Last Frame exposes the keyframe-conditioned video mode of FLUX 3, Black Forest Labs' multimodal generation model spanning image, video and audio in a single system. Rather than one conditioning image, you provide two anchor frames — the start and the end — and the model synthesizes the motion that connects them, which suits match cuts, product turns and storyboard-driven shot continuity.

The generational change is the main story. The FLUX 2 models in this catalog — Flux 2 Pro, Flux 2 Max and the editing variant Flux 2 Max — are still-image generation and editing systems. FLUX 3 is the generation in the line that produces motion, and Black Forest Labs describes its video output as arriving with natively synchronized audio, with clips chainable into longer multi-shot sequences. The lab also reports that FLUX 3 improves on earlier FLUX versions at handling complex prompts and rendering text.

Within the FLUX 3 video group, this endpoint sits between Flux 3, which works from a prompt alone, and Flux 3, which animates a single supplied image. Fixing both ends of the clip trades some generative freedom for tighter control over timing and framing.

Caveats worth noting: fine-grained specifications such as native resolution, frame rate and maximum clip length are not stated in this catalog entry, and the quality comparisons against earlier FLUX generations are the vendor's own.

This About section is AI-generated from public sources (Claude Opus 5), with no human editing. It may contain inaccuracies — verify critical details against the sources listed above.

Data sources: Venice API · HuggingFace · Wikipedia — enrichment updated 3h ago