phdelightful2 hours ago
My understanding (perhaps outdated) is that manufacturing variability is a key challenge for analog computing. Digital designs are also fundamentally analogue under the hood, but if you only need to resolve a 0 or 1 you are much more tolerant of any source of noise. I wouldn't mind hearing even a little bit more from Mythic about how they make this work in practice.
A 2026 EE Times article [1] refers to "compensation" and "calibration" techniques.
[1] https://www.eetimes.com/mythic-rises-from-the-ashes-with-125...
speps15 minutes ago
You reminded me the anecdote about every SID chip sounding different. If you hear a recording of a C64 made song, it’s unique to that chip (and somewhat to the machine as well, timing, crystal, etc.).
trebligdivadan hour ago
I'd assume it does some type of calibrate per device (regularly?) or the design is such that it's differential so things cancel out. Note it's also on 28nm for the analogue bits because yes it's harder.
MichaelNolanan hour ago
If you’re looking for their LLM page it’s https://www.mythic.ai/enterprise-llm
I wish they would have done what Taalas did with chatjimmy.ai and just directly host a model for us to view, rather than just claiming it’s 50x faster than Nvidia/groq. Their claim is specifically for a 1 trillion param model. So they could have just grabbed GLM 5.2, or similar, and hosted it.
vatsachak35 minutes ago
If they can't demonstrate it publicly it's probably fake.
mdp202141 minutes ago
> Mythic M1 stores up to 80 million neural network weight parameters directly on-chip
Which means connecting over 30 chiplets to run a Qwen 3.8 27b and over 3000 chiplets to run Qwen3.8-2.4T-A95B. Cost? Space? Feasibility?
tancopan hour ago
Their numbers look too good to be true, they have no identified customers and the whole site is generated, but I think the principle behind it is good. If they can pull off the error correction needed to make analog reliable we might have a great new option for cheaper more eco friendly AI. Then again it could turn out to be a total scam.
mdp202129 minutes ago
> Their numbers look too good to be true
Why? I have not seen anything outlandish for a NN implementation (vs a NN simulation).
> can pull off the error correction
There lie the issues that have not been mentioned, the solutions not explained. Analog computing means: * costly digital-to-analog at the input and analog-to-digital at the output; * sensitivity to environmental conditions such as temperature; * signal dispersion hence the need to boost it in the path.
Maybe checking the patents they registered?
vatsachak30 minutes ago
Like the numbers they claim could literally make LLMs 20x profitable. If it were true then why isn't every AI company trying to buy them out?
mdp20212 minutes ago
Notice that they do not talk about SRAM when they present the M1 chiplets - but you have to store the kv-cache etc. somewhere to run LLMs.
The technology that could run LLMs should be the "Vanguard", but as the homepage says, "the M1 (scope: Edge/Cameras/Drones) is there, the Vanguard should be a reality in 2027".
refulgentis22 minutes ago
c.f. https://news.ycombinator.com/item?id=49403836, then from there, you'd need to see a couple of orders of magnitude before it's tractable for LLMs.
Bottom of page linked from HN (currently https://www.mythic.ai/) indicates they're hoping to demonstrate something that could that in 2028 or later, and both Nvidia and Cerebra are looking at 10x'ing models to 10T+ plus in 2027.
So they may never catch up on LLMs.
They're a good fit for the companies they're working with and have taken investment from, ex. Toyota, that aren't doing LLMs.
vatsachak14 minutes ago
They plan to be able to run 1T parameter models next year
https://www.mythic.ai/vanguard
Seems big, IF true
refulgentis25 minutes ago
My 15 second read of just the front page aligned with you, but when I saw replies pushing back, I went back and loaded News, then cross-verified some of the claims. It's real.
SkyPuncher4 minutes ago
[dead]
alex7o2 hours ago
This looks cool a chiplet can fit 30m params so the biggest card can fit qwen 3.8 27b it would be cool to see some benchmarks on things like that publically.