Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on LiveCodeBench Hard

Re:Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on ... by ashkankiani on Saturday October 31, 2026 @09:37PM (#50321979)

It can run on one strix halo machine, which is what I have. For everything else, there’s a vibrant community brute forcing more efficient engines and quantizations. There’s a minimum bar right now, so it depends on your hardware.

[ Parent | on HN ]

Re:Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on ... by fooblaster on Saturday October 31, 2026 @09:58PM (#50322080)

you have 5k$ of ram in a strix halo?

[ Parent | on HN ]

Re:Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on ... by ashkankiani on Saturday October 31, 2026 @10:12PM (#50322141)

I bought my box for $3k last year because I hadn't updated my PC since the first threadripper 1950x, and the Strix Halo is the only x86 with unified memory that's optimized for the kind of CPU intensive stuff I do, and they finally have proper AVX512 too. It was the perfect upgrade for me (except for lack of PCIe 5 & ECC), which coincidentally happened to be perfectly positioned for local inferrence.

and I liked it so much (and saw that the RAM drought was going to last for at least a few years) so I bought a second one to use as a general server for myself and my family.

[ Parent | on HN ]