Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Qwen3.8-Flash-Next

qwen.ai

Post reply on HN
Print view View original

191–200 of 246 posts

Re: Qwen3.8-Flash-Next

#191

Post by flakiness »

mistersquid wrote: post #124
redrix wrote: post #53

Earlier quoted context omitted.

You’re absolutely right to be hopeful. Three honest possibilities, and I’ll be straight with you about each: 1. It overthinks — Just like the previous iteration. High confidence. 2. It doesn’t overthink — Improvement from the last model for your use case. Regression for others. 3. It sometimes overthinks — Best case all around. A feature, not an impairment. One final thing worth mentioning: (I made myself irrationall…

> You’re absolutely right to be hopeful. Three honest possibilities, and I’ll be straight with you about each: > [UGC styled humorously as LLMisms] All joking aside, having interacted with Claude intensely for the last 8 months and about 30 hours/week in the last 3, I’ve started to notice how (for want of a better word) “readable” (“digestible” ? “comprehensible” ? “Predictable” is the wrong direction.) information c…

I'm not sure it's a bad thing.

If you spend a long time with C++ code base you'll be able to decipher the otherwise-unreadable compiler errors pretty quickly, and I'd consider it a skill.

Re: Qwen3.8-Flash-Next

#192

Post by Giefo6ah »

peterleiser wrote: post #97
redrix wrote: post #53

Earlier quoted context omitted.

You’re absolutely right to be hopeful. Three honest possibilities, and I’ll be straight with you about each: 1. It overthinks — Just like the previous iteration. High confidence. 2. It doesn’t overthink — Improvement from the last model for your use case. Regression for others. 3. It sometimes overthinks — Best case all around. A feature, not an impairment. One final thing worth mentioning: (I made myself irrationall…

> Three honest possibilities, and I’ll be straight with you about each This. I don't know if the "honest answer" phrasing is part of the system prompt or alignment, but when people say "honestly" all the time I start wondering how honest they're being.

https://www.anthropic.com/constitution

Ctrl-F honest

57 matches

Re: Qwen3.8-Flash-Next

#193

Post by GrinningFool »

a_humean wrote: post #36

Waiting for llama.cpp support to land, but this might be a big deal for Strix Halo users. 6B active params helps around the memory bandwidth constraints, but a 128GB box can probably run the Q3/Q4 quants fairly easily with a decent context size. This might actually be better for strix users than 27B, which was already very good.

Using llama.cpp I one-shotted (2 hours) a reasonable asteroids clone on my strix halo/128 using the 1 bit quant, using my custom harness (which isn't anything exceptional).

It was tedious - a lot of second guessing itself, and quadruple checking things it fixed a couple of iterations back - but it got there and the result is a playable game.

Speed starts out strong, but definitely drops off as context grows. At the end (I think context about 70k) it was down to 12 output tps.

Mind a bit blown.

Re: Qwen3.8-Flash-Next

#194

Post by mdp2021 »

hedgehog wrote: post #162
mdp2021 wrote: post #154

Earlier quoted context omitted.

> Better... than 3.8 27B How can a "6b active per token" MoE be better than a just released 27b in the same family? Possible maybe, but quite interesting and requiring some explanation. Edit: ok, on second thought, probably because the small "experts" are really rich in specialized training compared to the dense model. Still raising questions about the details, e.g. the reasoning abilities (or all meta-skills) of a "…

Separating knowledge from reasoning so you only pay for what you use is a big rationale for MoE, the big problem being MoE training has historically been hard to get right. In a dense model every single token you're paying a cost to determine whether you're now talking about the flavor of durian.

Sure, the "6b subset" can be more knowledgeable on its area than a whole 27b generalist (and more efficient), but where is the simulated Intelligence encoded? A 6b subset as or more intelligent than a 27b raises the question of how metacognition skills are stored.

Re: Qwen3.8-Flash-Next

#195

Post by NamlchakKhandro »

pixelpoet wrote: post #117
samsari wrote: post #85

Earlier quoted context omitted.

70% of the posts on HN are already satire and performance art

And full of made up statistics.

Which then devolve into supporting arguments for communism. Themselves becoming food for future irrational anecdotes about communism.

Re: Qwen3.8-Flash-Next

#197

Post by pyaamb »

simonw wrote: post #126

I ran some pelicans at the four different reasoning levels (none, low, medium, xhigh - apparently high and xhigh are aliases of each other) on a DGX Spark using Unsloth's unsloth/Qwen3.8-Flash-Next-GGUF (UD-IQ1_S): https://tools.simonwillison.net/markdown-svg-renderer#url=ht... Surprised I didn't get one I liked as much as the Qwen 3.8 27B one https://simonwillison.net/2026/Aug/16/qwen-38-27b/#the-defau... , maybe be…

The Pelican Brief

Re: Qwen3.8-Flash-Next

#198

Post by phatfish »

Giefo6ah wrote: post #192
peterleiser wrote: post #97

Earlier quoted context omitted.

> Three honest possibilities, and I’ll be straight with you about each This. I don't know if the "honest answer" phrasing is part of the system prompt or alignment, but when people say "honestly" all the time I start wondering how honest they're being.

https://www.anthropic.com/constitution Ctrl-F honest 57 matches

Haha, that is like the wall of nonsense text that used to be hidden on link-farm pages for SEO. The "final word" is delusional.

Re: Qwen3.8-Flash-Next

#199

Post by potus_kushner »

simonw wrote: post #173
simonw wrote: post #126

I ran some pelicans at the four different reasoning levels (none, low, medium, xhigh - apparently high and xhigh are aliases of each other) on a DGX Spark using Unsloth's unsloth/Qwen3.8-Flash-Next-GGUF (UD-IQ1_S): https://tools.simonwillison.net/markdown-svg-renderer#url=ht... Surprised I didn't get one I liked as much as the Qwen 3.8 27B one https://simonwillison.net/2026/Aug/16/qwen-38-27b/#the-defau... , maybe be…

Tried again with a different quant, UD-Q2_K_XL: https://tools.simonwillison.net/markdown-svg-renderer#url=ht...

the xhigh version looks amazingly good for a 2 bit quant.

Re: Qwen3.8-Flash-Next

#200

Post by potus_kushner »

at first this looked like something one could run on CPU with 64GB RAM with a 2-3 bit quant, at possibly half the speed of 3.6 35B-A3B, however the 50B ngram sidecar makes it impossible. and oddly, unsloth's page lists the ngrams as 50GB even though they say it's in 4 bits. should be 25GB according to my math. anyway, the new ngram architecture makes it pretty much unusable for regular folks who cant afford more than 32-64 GB ram in this RAMocalypse.
Post reply on HN