@FinanceYF5: Running this 24-hour uninterrupted AI livestream with fal costs about $3,450 per day; switching to Seedance, the same c…
Summary
The article compares the daily costs of AI livestreaming using fal versus Seedance, highlighting that fal is about 10 times cheaper and ranks highly in text-to-video performance.
Similar Articles
Been on the Seedance 2.5 API all morning. The cost structure is the part worth knowing.
Seedance 2.5 launched, enabling 30-second single-pass video generation, audio-reference lip-sync, and API features like async jobs and webhooks. Its standout is billing for failed generations isn't charged and reference videos get lower token rates, though output max is 720p.
@FinanceYF5: fal, Silicon Valley’s AI image-generation inference cloud, has $800 million in annual revenue and is seeking a $20 billion valuation. What justifies that? Two years ago, it was making only $10 million a year. fal isn’t selling GPUs—it’s selling a lineup of models that can be swapped out at any time. A different video model takes the top spot every month; there’s no winner. fal makes its money precisely because there’s no winner. Four questions: What’s different?…
The article analyzes the rapid rise of fal, a Silicon Valley cloud inference platform for AI image and video generation: its annual revenue is approximately $800 million, and its valuation could reach $15–20 billion, compared with annual revenue of only about $10 million two years ago. Rather than selling GPUs, fal’s core business model is to offer a lineup of models that users can switch between at any time, profiting from a market where models evolve rapidly and no winner stays on top for long.
@VraserX: Seedance 2.5 making consistent videos up to 3 minutes in one go is honestly insane. If this pace continues, we’ll be ma…
Seedance 2.5 enables consistent AI-generated videos up to 3 minutes in one go, marking a significant leap in AI video generation that could soon lead to TV episodes and feature films.
@EMostaque: So @deepseek_ai v4.1 Flash I estimate cost $10m to train, 100x less than GPT-6 Astra. It also costs 100x less to run an…
A tweet estimates that DeepSeek AI's v4.1 Flash model cost $10 million to train, 100 times less than GPT-6 Astra, with lower running costs and similar benchmark performance, prompting questions about the bull case for frontier AI labs.
@FinanceYF5: The video industry has also seen the "Jevons Paradox": the lower the production costs, the more content consumption the…
AI has reduced the production cost of animated short dramas to 1/10 and increased speed by 450 times, leading to a 12-fold growth in output on Douyin, with 89% of top dramas produced by AI.