Fal · mars 2026
11 mises à jour · mars 2026
L'archive Spyingbee contient 11 pages suivies pour Fal, enregistrées pour la première fois en mars 2026. Elles ne portent pas de date de publication : la date indiquée est celle de notre première collecte. Le thème dominant : pricing updates (3). À retenir : « fal-ai/kling-video/o3/standard/image-to-video » (30 mars 2026).
Parcourir d'autres mois de Fal
Fal.ai has introduced a new image-to-video generation model, fal-ai/kling-video/o3/standard, which converts static images into videos. The model charges users per second of generated video, with prici…

Fal.ai has launched a new version of its image-to-video model, kling-video/v3/pro, which enables users to convert static images into videos with customizable controls. The tool accepts multiple image …

Fal.ai has introduced FLUX.2 [flex], a customizable text-to-image model designed for professional-grade typography and precise control over image generation. The model allows users to adjust inference…

Fal.ai has upgraded its Nano Banana image-editing model to Nano Banana 2 (officially Gemini 3 Pro Image), replacing the original Nano Banana (Gemini 2.5 Flash Image). The new model costs $0.15 per edi…

Fal.ai has introduced a new feature for its Veo 3.1 model, enabling users to generate video from just the first and last frames of an image sequence. This innovation simplifies the video creation proc…

Fal.ai has launched a new service called FLUX.1 LoRA Fast Training, enabling developers to train custom Low-Rank Adaptation (LoRA) models on top of the FLUX.1 base model at unprecedented speeds. The s…
Fal.ai has launched Nano Banana 2, a new text-to-image model built on Google’s Gemini 3.1 Flash Image architecture, offering faster generation speeds and improved output quality compared to its predec…

Fal.ai has launched Inworld TTS-1.5 Max, a new text-to-speech (TTS) model optimized for low-latency, expressive, and multilingual voice synthesis. The model achieves time-to-first-audio under 250ms, s…
Fal.ai has launched the fal MCP Server, a hosted endpoint enabling AI assistants to directly search, run, and chain over 1,000 generative AI models from a single conversation without requiring SDKs or…

Fal.ai announced an invite-only evening event focused on real-time AI across industries, emphasizing realtime world models and conversational AI. The event is restricted to attendees 21+ and requires …
HeyGen's models are now available on the fal platform, offering capabilities such as talking avatars, automated video generation, and multilingual lip sync. The Video Agent API allows users to create …
Suivez Fal en pilote automatique