Google just mogged this tame impala test.
๐ I gave 4 fast models the same Currents cover. Same prompt. One shot via aimlapi. โข Gemini 3.8 Flash 2:58 โข DeepSeek V4.1 Flash 2:54 โข Muse Spark 1.3 3:34 โข Qwen 3.8 Omni-Flash 3:50 Four completely different interpretations.
Constantly amazed at what you can do at home with *significant* investment (but also not boat money).
My own agent/harness running entirely on my own hardware. It's the Qwen 3.8 Flash model running on one Spark, plus all the voice models running on a 3090. It's plugged into OMP,
The best looking result and the best running result now come from two different models.
We got used to quality having one axis and one winner. One prompt, a night FPV flight over a skatepark: GLM-5.3: visually the best, but 655 s and 118k tokens, $0.52. GPT-6 Astra: clearly
ใWhat is the truth of this clip?ใ
Rolling A popular clip with the following comment: ููููููููููููููููููููููููููููููููููููููููููููููููููููููููููููููููููููููููููู It is not an AI movie, but perhaps science fiction, and it may come true in 2040 โข China officially unveils the Luan Niao, the largest aircraft carrier on the planet โข Size: equivalent to 7 football fields โขโฆ
You have reached the end of the archive
All of deepseek