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

Qwen 3.8 27B

huggingface.co

Post reply on HN
Print view View original

11–20 of 848 posts

Re: Qwen 3.8 27B

#11

Post by edg5000 »

scrlk wrote: post #8

Beats Opus 4.7 Max (w/ Claude Code) on DeepSWE (42.2 vs 40). Looks like Qwen's 27B models continue to pack some punch. Unsloth's GGUF quants are up: https://huggingface.co/unsloth/Qwen3.8-27B-GGUF

That's crazy, considering the massive size difference. But the small Qwen models are known for punching above their weight.

Re: Qwen 3.8 27B

#14

Post by nblgbg »

scrlk wrote: post #8

Beats Opus 4.7 Max (w/ Claude Code) on DeepSWE (42.2 vs 40). Looks like Qwen's 27B models continue to pack some punch. Unsloth's GGUF quants are up: https://huggingface.co/unsloth/Qwen3.8-27B-GGUF

Is there any advantage to using the model from Unsloth compared with https://huggingface.co/Qwen/Qwen3.8-27B-FP8 ?

Re: Qwen 3.8 27B

#15

Post by m_ke »

kristopolous wrote: post #9

q4km is about 48 tps on a 4090. my llama.cpp params are --flash-attn on --parallel 1 --load-mode mmap

With spec decode should easily get to >100tps

on my dual 3090s qwen 3.5 27b was running at around 110tps using the config from https://github.com/noonghunna/club-3090

make that 200tps on a single 5090, 4x faster than opus https://x.com/radixark/status/2088285681131110446

devs about to get handed a two 5090 box each and told to max that out

Re: Qwen 3.8 27B

#16

Post by benxh »

nblgbg wrote: post #14
scrlk wrote: post #8

Beats Opus 4.7 Max (w/ Claude Code) on DeepSWE (42.2 vs 40). Looks like Qwen's 27B models continue to pack some punch. Unsloth's GGUF quants are up: https://huggingface.co/unsloth/Qwen3.8-27B-GGUF

Is there any advantage to using the model from Unsloth compared with https://huggingface.co/Qwen/Qwen3.8-27B-FP8 ?

Depends on what software/hardware you'll run it. GGUFs from Unsloth can run on pretty much every single potato; full weights need beefy gpus

Re: Qwen 3.8 27B

#20

Post by NorwegianDude »

If the benchmarks are a real indication, we now have a local model that is runnable on a high-end personal PC that trades blows with the leading model Claude Opus 4.6 Max from half a year ago.

Insane if that is the case. Downloading now!

Post reply on HN