News for hackers. Stuff that matters.
Re:AMD acquires Taalas to boost inference performance by etc... by nojs on Thursday August 06, 2026 @05:43PM (#49202996)
Can anyone comment on the economics and likely turnaround times of this process, when it’s more mature?

Would it be realistic for a frontier lab to deploy this or would the turnaround time mean the model is always too out of date?

Assuming the weights and architecture are eventually stable, how much cheaper would this end up being?

Re:AMD acquires Taalas to boost inference performance by etc... by 2001zhaozhao on Thursday August 06, 2026 @05:49PM (#49203067)
There are always uses for outdated models.

Claude Code is still using haiku 4.5 from ages ago for explore subagents for instance. Not to mention production uses like customer service that only need to be "good enough"

Re:AMD acquires Taalas to boost inference performance by etc... by edot on Thursday August 06, 2026 @05:56PM (#49203169)
Just looked this up, no longer true. Explore subagents inherit whatever model the parent is. And you can of course make other subagent configs.
Re:AMD acquires Taalas to boost inference performance by etc... by samtheprogram on Thursday August 06, 2026 @06:04PM (#49203240)
That's solely so that you burn more money. It's totally unnecessary to assume the parent model. Sure, it could be upgraded from Haiku if there was a solid reason to, but...
Re:AMD acquires Taalas to boost inference performance by etc... by AussieWog93 on Thursday August 06, 2026 @06:06PM (#49203264)
I mean, if you could get Opus or even Sonnet 4.5 at 1000+ tok/s exploring the codebase, they would probably change that setting back.

But either way, I think GP's overall sentiment of "delegating intelligence-saturated tasks to an outdated but fast subagent" makes a lot of sense.

Re:AMD acquires Taalas to boost inference performance by etc... by alightsoul on Thursday August 06, 2026 @06:12PM (#49203338)
Customer service has really degraded huh. 4 years ago they expected opus performance out of human call center agents

I guess losing some customers due to poor customer service is ok if the price of customer service is right.

Re:AMD acquires Taalas to boost inference performance by etc... by cogman10 on Thursday August 06, 2026 @06:09PM (#49203290)
2 to 3 months optimistically assuming everything goes smoothly and is fully automated.

6 months or even a year if something goes wrong in the fabrication process and you need to update things.

If they do more standard asic design, it could be a lot longer as the design needs to be validated on an FPGA cluster, which would necessarily need to be very big for something like a LLM. Easily up to 2 years.

There's a reason chatjimmy isn't demonstrating newer models and why they only show of an 8B model.

Re:AMD acquires Taalas to boost inference performance by etc... by shangofox on Thursday August 06, 2026 @05:50PM (#49203075)
I mean even if it take a few months, it'll still be out of date. But there was a hypothetical when it came up in Feb, would you want Qwen 3.5 at like 10k tokens per second.

At the time people were no doubt saying yes but now 3.8 is out, is that still desirable?

Re:AMD acquires Taalas to boost inference performance by etc... by xienze on Thursday August 06, 2026 @06:06PM (#49203267)
There's soooo much stuff that such a model is still capable of doing in the pursuit of getting a better overall answer. Imagine a powerful research agent that blasts out dozens of the small, cheap models to fetch and summarize one page each. Then the beefy researcher model performs the final analysis.