@MiaAI_lab I asked GLM 5.3 Flash if it was trained in financial analysis and modeling. It said it had a PhD level as a quant. I asked deepseek flash 4.1 and even after 3 prompts (and 3 times the cost) I'm still not sure if it gave me a straight answer.
@Zai_org So not using Nvidia chips, huge week of free usage and hype and release tonight, which happens to be when Nvidia is releasing earnings. Coincidence?๐ค
@ashen_one My quibble is, if all of your other points become true, will RAM prices really continue to increase at a higher rate? With increased efficiency, shouldn't RAM prices stabilize? Unless way more people start using LLMs. Just not convinced that's going to happen either.
@ratujma@SourceCodeplz@EpochAIResearch Better than GLM 5.2 as an advisor? Asking because I've been trying to see for myself which is better to use with Deepseek Flash.