Deepseek flash made this teaser video one shot.
Consumed 2% of my Opencode go 5hr session. The script is vague, but the sounds and visuals - loving it! Tried the same with a SaaS launch video - was not so good. But I guess the output also depends on the kind of input. In
I gave all three of my local models the same prompt to build a neon rain alley in three.js.
GLM 5.3 Flash ran at 95 tok/s and DeepSeek V4.1 Flash at 127 on my 4 box Spark ring, and Qwen 3.8 27B did 42 on my Mac Studio while it was busy with other work. Which one built it best?
FREE AI: DeepSeek, Claude Haiku, GPT-6 Luna and 1M context Inkling, all at $0 this week
CanTrustAI: free models with no monthly limit, including DeepSeek V4.1 Flash, Claude Haiku 5.5, GPT-6 Luna and Seedream. paid models at 50-75% off. 1,000 welcome credits with Google:
DeepSeek with all the signs of burnout and ignoring 🤡🤡 NOBODY will understand but that's ok, right, the things I invent help
You have reached the end of the archive
All of deepseek