ElevenLabs releases SFX v3, a sound effects model that follows multi-part prompts more closely
ElevenLabs released SFX v3 in its Sound Effects API. Compared with SFX v2 it follows prompts with several instructions more closely, handles vocals and music better and clips less on high sounds such as gongs and chimes. Set model_id to eleven_text_to_sound_v3 to use it; looping still works only with v2.
Why it matters
Both models share one endpoint, so developers can run their usual prompts through v2 and v3 by changing one parameter before deciding to switch.
Sources1
Written by dalao.ai: the title, summary and why it matters are our own words. The original reports are the record; their publishers keep authorship.