Honestly, this is starting to make more and more sense. SOTA models are starting to converge to certain architecture and capabilities. I wouldn’t be surprised we end up with a base model ASIC + “fine tune” card where it’s a physical LoRA style adapter.
AMD acquires Taalas to boost inference performance by etching models in silicon
11–20 of 712 posts
Re: AMD acquires Taalas to boost inference performance by etching models in silicon
#12Toronto Canada startup btw.
Re: AMD acquires Taalas to boost inference performance by etching models in silicon
#13Honestly, this is starting to make more and more sense. SOTA models are starting to converge to certain architecture and capabilities. I wouldn’t be surprised we end up with a base model ASIC + “fine tune” card where it’s a physical LoRA style adapter.
The technical aspects of SOTA models are not publicly documented. How do you know if something is converging?
Re: AMD acquires Taalas to boost inference performance by etching models in silicon
#14Honestly, this is starting to make more and more sense. SOTA models are starting to converge to certain architecture and capabilities. I wouldn’t be surprised we end up with a base model ASIC + “fine tune” card where it’s a physical LoRA style adapter.
The technical aspects of SOTA models are not publicly documented. How do you know if something is converging?
if they were still exponentially increasing, they wouldn't be preparing for an IPO. IPO is where companies go to die and founders escape.
Re: AMD acquires Taalas to boost inference performance by etching models in silicon
#15Honestly, this is starting to make more and more sense. SOTA models are starting to converge to certain architecture and capabilities. I wouldn’t be surprised we end up with a base model ASIC + “fine tune” card where it’s a physical LoRA style adapter.
The technical aspects of SOTA models are not publicly documented. How do you know if something is converging?
If we had deepseek v4 flash 0731 etched on a chip it would be more than capable enough and fast enough for so many people's needs, even hardcore engineer.
Re: AMD acquires Taalas to boost inference performance by etching models in silicon
#16Honestly, this is starting to make more and more sense. SOTA models are starting to converge to certain architecture and capabilities. I wouldn’t be surprised we end up with a base model ASIC + “fine tune” card where it’s a physical LoRA style adapter.
I don't think there'll be a fine tune card; you'll have the base model vintage whatever year, and then your GPU will do whatever LoRA layers you want it to do; the LoRA will wrangle older dated models into the current of whatever your looking at.
But yeah, for things like programming, if it can do linux and python and some go and sql and javascript, larger domains can be threaded with LORA
Re: AMD acquires Taalas to boost inference performance by etching models in silicon
#17The demo: https://chatjimmy.ai/
It doesn’t believe it’s running on that chip, it’s arguing with me
Re: AMD acquires Taalas to boost inference performance by etching models in silicon
#18Honestly, this is starting to make more and more sense. SOTA models are starting to converge to certain architecture and capabilities. I wouldn’t be surprised we end up with a base model ASIC + “fine tune” card where it’s a physical LoRA style adapter.
The technical aspects of SOTA models are not publicly documented. How do you know if something is converging?
SOTA American models are not. SOTA Chinese models are. From a physics aspect, closed source models cannot be too far from open source ones in terms of size. There’s only so much you can squeeze out a B100 style cluster even with fancy Dflash style diffusion model for the speculative model.
Re: AMD acquires Taalas to boost inference performance by etching models in silicon
#19Honestly, this is starting to make more and more sense. SOTA models are starting to converge to certain architecture and capabilities. I wouldn’t be surprised we end up with a base model ASIC + “fine tune” card where it’s a physical LoRA style adapter.
Wouldn't this mean someone with sufficient hardware could lift the SOTA model weights off the chip? Or are you saying that these chips would only be used internally by these companies and not sold to the public?
Re: AMD acquires Taalas to boost inference performance by etching models in silicon
#20Honestly, this is starting to make more and more sense. SOTA models are starting to converge to certain architecture and capabilities. I wouldn’t be surprised we end up with a base model ASIC + “fine tune” card where it’s a physical LoRA style adapter.
Wouldn't this mean someone with sufficient hardware could lift the SOTA model weights off the chip? Or are you saying that these chips would only be used internally by these companies and not sold to the public?
I don’t get why this is an issue? You can run Claude/OpenAI SOTA models through Amazon bedrock. These weights have to live somewhere to run on Bedrock.