Skip to main content

fal News

Kandinsky 6 models arrive on fal

October 6, 2026

fal added Kandinsky 6 Pro and Lite video models with natively synchronized audio, alongside VSR and VSR Lite video super-resolution models.

fal creative generation plugin for ChatGPT and Codex

October 6, 2026

The fal plugin brings image, video, audio and 3D generation to ChatGPT Work and Codex. Start from a prompt or reference image, choose a model with the assistant and refine results through editing and conversation. Access depends on plan and workspace settings.

Nano Banana 2.1

October 6, 2026

via Higgsfield

Google's update to Nano Banana 2 is now available on Higgsfield, with the same speed and output up to 4K. Compared with Nano Banana 2, it improves visual quality, prompt adherence, character consistency across edits, and text rendering, per Google. References: up to 14 images, with consistency for up to 4 characters and up to 10 objects. Text edits: describe what to change, such as a shirt color, and the model edits that part of the image. Wide formats: fixed tiling artifacts on 1:4, 4:1, 1:8, and 8:1 at 2K and 4K. Resolution: 1K by default, 2K, or 4K. To get started: Image → model selector → Nano Banana 2.1. Nano Banana 2.1 is also available in Google Flow, ElevenCreative, fal, Magnific, OpenArt, Pika and Krea. FLORA announced Google’s latest image model; Weave announced a new Nano Banana model with a temporary 50% credit discount, without naming the version in those posts.

Gemini 3.8 Flash TTS Models on fal

September 25, 2026

fal added Gemini 3.8 Flash TTS and Flash Lite TTS for expressive speech generation. The models support voice and delivery direction through text, including pacing, emotion, dialect, and conversational cues.

Recraft V4.1 Flash on fal

September 23, 2026

Recraft V4.1 Flash is now available on fal. The image model generates in about 1.3 seconds, making it quicker to test compositions, revise prompts, and explore new creative directions.

3D to Video on H3 Max

September 23, 2026

fal added 3D to Video for H3 Max. Upload a Blender previs up to 15 seconds long to turn it into photoreal video while preserving camera movement and shot timing. Reference images can guide the final shot’s characters and environments.

State of Generative Media Report Volume 2

September 22, 2026

fal released the second volume of its State of Generative Media report, drawing on its platform data to examine how AI image, video, audio, and 3D are moving into production. The accompanying thread highlights video editing and shot extension, faster creative iteration, and a shift from individual assets toward objects, environments, and experiences. fal Design Labs created print and interactive digital editions.

Meshy V7.1 on fal

September 21, 2026

fal added Meshy V7.1 for generating textured 3D assets from text or image prompts. It can combine multiple views of an object into one model and generate optional PBR maps for realistic materials and lighting.

H3 Max Styles on fal

September 21, 2026

fal added built-in visual styles for H3 Max video generation: 16-bit pixel art, hand-drawn, low-poly, 1970s retro toon, and VHS. Each style guides textures, color, motion, and visual details, and can be used with a text prompt or a reference image.

Tripo3D P2 on fal

September 21, 2026

Tripo3D P2 is now available on fal for generating textured 3D meshes from text prompts or images. It offers four texture-quality settings, separate seeds for geometry and textures, and an option to remove baked-in lighting when using the v3.5 texture model.

Lyria 3.5 Music Generation on fal

September 19, 2026

Google DeepMind’s Lyria 3.5 is now available on fal. It turns detailed text prompts into music, blends genres and instrumental styles, and lets creators direct vocal delivery and musical dynamics through text.

Grok Voice Is Live on fal

September 16, 2026

Grok Voice, the latest speech model from SpaceXAI, is live on fal. It answers in 0.70 seconds and finishes its tool calls before the sentence ends, with transcription that includes word-level timestamps, text to speech in 30+ voices across 25+ languages, and voice cloning from two minutes of audio.

Boreal by Creatify Is Now on fal

September 15, 2026

fal has added Boreal, a video generation model from Creatify. It turns a prompt into product ads, UGC, and presenter videos with synchronized speech, music, and sound effects. Generation can be guided with an image, audio, or both, with clips up to 20 seconds and 2K resolution.

fal Agent Adds $50 Per Month Tier

September 15, 2026

fal Agent now has a $50 per month tier for people who want to experiment. With H3 Max, a 5-second 768p video costs $0.20 and generates in under 3 seconds. fal Agent is one workspace for every image, video, and 3D model on fal, choosing the right model for each task while keeping characters and references consistent across models. The full $50 comes back as fal credits usable across Agent, Playground, and anywhere on fal, and unspent credits roll over.

H3 Max and H3 Max Turbo 50% Off on fal Through September 30

September 15, 2026

fal has cut the price of H3 Max and H3 Max Turbo by 50% through September 30. During the sale, H3 Max at 768p costs $0.04 per second and H3 Max Turbo at 768p costs $0.02 per second. fal describes them as the fastest frontier video models on the market.

Billing Headers for WebSocket Endpoints

September 14, 2026

Shared WebSocket endpoints can now declare billing on their 101 upgrade response. Set x-fal-billable-units when the unit count is known at connect time. Set x-fal-billable-units-webhook to 1 when the count is only known at the end of the session. Then report the total with POST /requests/billable-units/{request_id}. The platform still charges sessions without either header by duration. Learn more about billable units

FLUX 3 Video Edit on fal

September 10, 2026

FLUX 3 Video Edit is now live on fal. Use one prompt to swap characters, rebuild backgrounds, restyle shots, rewrite dialogue, change colors and materials, or alter a clip’s events. Lip sync follows new dialogue automatically, including translations.

Dreamina Seedance 2.0 on fal

September 10, 2026

Dreamina Seedance 2.0 is now available to US companies on fal, hosted entirely on US infrastructure. It brings native audio-video generation and multimodal control across text, images, video, and audio for complex, production-grade video workflows.

Switch Between App Endpoints in the Dashboard Playground

September 9, 2026

Deployed Serverless apps now include Testing > Playground in the app dashboard. Select an available endpoint, fill in its generated input form, and run a request without leaving the app. Multi-endpoint apps have one place to test their endpoints, using the existing Playground authentication and billing behavior. Open the guide

H3 Max Multi-angle on fal

September 9, 2026

Multi-angle is now live for H3 Max on fal. Set horizontal and vertical camera angles in degrees and move through a scene from a single viewpoint in under three seconds, while shape, position, and materials stay consistent across perspectives.

Ad Delayer from Bria on fal

September 9, 2026

Ad Delayer from Bria is now live on fal. Recover a finished flat ad as a layered, editable composition; headlines return as SVG text for localization, and layers arrive typed and ordered.

H3 Max Director 1.1 on fal

September 8, 2026

H3 Max Director 1.1 is now available on fal with keyframes throughout a generation, native audio conditioning, native 1080p output, and continuous generations up to 15 minutes. Director adapts to new visual and audio inputs in real time while preserving continuity.

H3 Max Turbo (preview)

September 2, 2026

H3 Max Turbo is a faster, more cost-efficient version of H3 Max: 2x the speed at half the cost, targeting the 97th percentile of H3 Max quality on fal's evals while still beating H3 and other video models. At the promotional price, 768p video runs at $0.01 per second of output. It keeps H3 Max's prompt understanding, aesthetics and overall quality while cutting generation time in half, and is available at the promotional price for two weeks.

Fibo Edit 1.5 on fal

September 2, 2026

Fibo Edit 1.5 from Bria is live on fal. It now accepts up to four reference images, so you supply the actual product, garment or character rather than describing it and the identity carries through. It is built for catalog and ad production: product placement, virtual try-on and consistent characters.

H3 Max Live

August 29, 2026

fal has H3 Max generating video faster than real time. The demo is an infinite broadcast where every frame is generated on the fly and each scene is directed from chat: type !prompt and it is on screen within seconds.

Gemini Omni Flash 1.1 on fal

August 27, 2026

Gemini Omni Flash 1.1 is on fal, processing text, image and video together rather than in sequence. It brings Gemini's real-world and physics knowledge to generation, and edits through conversation while holding character and scene consistent across turns.

Muse Image from Meta

August 26, 2026

Muse Image, Meta's agentic image model, is live on fal. It plans the composition, uses tools and self-corrects before returning anything, which is aimed at complex briefs that usually fall apart partway through.

H3 Max

August 26, 2026

H3 Max is a post-trained video model from fal Research. fal says it ranks first for overall quality, prompt understanding and aesthetics against leading video models on both its own and third-party independent evaluations, while generating a 5-second 720p video in under three seconds. It is 50% off for a week from launch.

Platform MCP Server

August 17, 2026

fal's second MCP server — the Platform API MCP — is now available at https://api.fal.ai/v1/mcp/platform. Where the Run MCP lets your AI assistant *build with models*, this one lets it *operate your fal account*: connect Claude Code, Cursor, or any MCP-compatible client with your API key and ask *"why did my last request to my-app fail?"* — your assistant walks the same debugging steps you would. 15 tools: eleven first-class serverless debugging tools covering requests, logs, error analytics, deploy and revision history, runner and queue state, spend, and persistent storage — plus a four-tool discovery gateway that opens up the rest of the Platform API (compute, workflows, keys, account, organization, storage) without bloating your assistant's context. bash theme={null} claude mcp add --transport http fal-platform \ https://api.fal.ai/v1/mcp/platform \ --header "Authorization: Key YOUR_FAL_KEY" Read-only — it can observe your account but never change it; write operations are listed in the catalog but refuse to execute Stateless — your key is sent per-request and never stored, and normal permission and ownership scoping applies Free — tool calls are the same Platform API calls you co

New Serverless Observability APIs

August 17, 2026

A batch of endpoints rounds out programmatic access to what the dashboard shows about your serverless apps: GET /v1/serverless/apps — list your deployed apps with live state: active runners, queue size, machine types, and environment. Add expand=endpoints to include each app's route-level endpoint ids, in the exact form the requests and analytics APIs accept. GET /v1/serverless/apps/{owner}/{name}/events — operational event history: deployments, configuration changes, and runner lifecycle transitions — the "what changed around this time?" API. Also surfaced in the dashboard's events timeline. GET /v1/serverless/apps/{owner}/{name}/revisions — revision history with deployment status, who deployed, and deploy messages and annotations (the API behind the August 3 entry above) — line revision boundaries up against your error timeline to spot a bad deploy. GET /v1/serverless/apps/{owner}/{name}/runners/history — historical runner counts by state (running, idle, pending, draining) for capacity analysis and incident forensics. Together with the existing requests, logs, analytics, metrics, and usage APIs, these power the new Platform MCP server — and they're plain REST endpoints you can ca

MiniMax H3 on fal

August 11, 2026

MiniMax H3 is now available on fal. The physics are insane: water that lands with real weight, cloth that clings once it's soaked. A slow push-in on the face holds up too.

Mirage Avatar X API on fal

August 11, 2026

The Mirage Avatar X API is live on fal. Build expressive avatar experiences with Mirage's most advanced generation model yet, with identity preservation across AI video. Get started with audio and an optional visual reference.

Introducing fal Agent

August 7, 2026

Following a week of frontier video model launches, fal is opening early access to something it's been building: fal Agent, a creative partner for working across the best image, video, and 3D models.

Seedance 2.5 on fal

August 7, 2026

Seedance 2.5 is live on fal. In fal's demo, the input is a still image with a red squiggle, and the output is a continuous FPV run through the torii gate with a full lap around the monorail. Nobody keyframed anything.

GPU Utilization in Runner Telemetry

August 4, 2026

Runner telemetry now includes a dedicated GPU Utilization chart alongside the existing CPU, memory, and VRAM charts, so you can see how much compute your GPUs are actually doing -- not just how much memory they hold. GPU Utilization reports the average compute utilization across a runner's GPUs over time, surfaced both as a current stat and a time-series chart in the runner side sheet. Pair it with VRAM Usage to tell a memory-bound runner apart from a compute-bound one, making it easier to right-size machine types and spot under-utilized GPUs. Open any runner from the Runners page (Dashboard → Apps → \[your-app] → Runners) to view its telemetry. See Runner Analytics for details.

Deploy Messages and Annotations

August 3, 2026

You can now attach a freeform message and custom annotations to each revision when deploying: bash theme={null} fal deploy my_app.py::MyApp --message "Deploying new version" --annotation GIT_SHA=a1b2c3d4 Messages and annotations are set at deploy time and stay with the revision. View and search them on the app's Versions page in the dashboard, or fetch them via the revisions API. Also available in the Python SDK: client.deploy(..., message=..., annotations={...}). See Annotating Deployments for details.

LTX-2.3 Reframe on fal

July 16, 2026

LTX-2.3 Reframe is now live on fal, from LTX. Convert any video to a new aspect ratio: landscape to vertical, square, and back (16:9, 9:16, 1:1, 4:5, 5:4). Newly exposed areas are generated to match the original scene, with up to 60 seconds of input and 720p or 1080p outputs.

Veed Lip Sync 2.0 on fal

July 14, 2026

Lip Sync 2.0 from VEED is now live on fal. Swap the audio on any video and automatically match the lips, emotion, style, and timing, no training or fine-tuning required. Process footage up to 4K, clips up to 10 minutes, with controllable sync strength.

Open-Sourcing 3DREAL

June 26, 2026

fal open-sourced 3DREAL — a render-to-real IC-LoRA for LTX-2.3 that turns any 3D, CG, or game render into photorealistic, film-quality video while keeping your exact composition and camera move from the first frame.

Seed Audio 1.0

June 25, 2026

ByteDance's Seed Audio 1.0 is now available on fal — generate natural speech, sound design, and music from text. Try it today.

LTX 2.3 Video Effects

June 25, 2026

A new set of LTX 2.3 video effects is live on fal — apply an effect to any video in a single pass with in-context effect LoRAs: water simulation, cross-eyed, day to night, colorization, deblur, decompression, and instant shave.

View fal on Creators Toolbox