There are so many of us with weak GPUs and these updates are massive wins. Still waiting for the 2080 moment where this 27B param model and a reasonable kv cache can exist and function at slow, but overnight-burning speeds ..
This would depend hugely on the price of electricity, however I'm pretty sure where I am (at roughly 25c/kwh in usd) the price of running qwen costs more than 6.1-luna per token. So other than privacy (which doesn't matter for my work) there isn't any justification for running locally. Is privacy purely what people justify local models on?
As an experiment I took, not this current quant, of Qwen 3.8 27B and had it analyze some of my writing in CPU-only mode with llamacpp. Tried a Q4 and Q5. It was dreadfully slow, but you fire that off before bed, and you can get results the next day. This was a ~6000 word markdown file, took maybe 10 minutes before it started generating anything at all, but then it just cranks away. I would need to tune down the default thinking intensity, but it otherwise did the job.
Not compatible with vibe coding, but if you want a local tool to review content, it is possible today.
I’ve been using Bonsai 2 (also based off of Qwen3.8 27B) on my mac mini in this way (using pizza-bot for the harness). In my case I’ve been using more for deep web search type stuff. It’s been a nice workflow to kick off a request in the evening and come back to a 1 page report on it the next morning.
Just installed Qwen 3.8 Splash tonight (doesn’t appear on linked page). Super impressed. Easily averaging 45 tps, even with thinking on. Totally acceptable. Running M4 Mac mini, 48GB, LM Studio.
Bonsai 2 does some additional tuning in addition to just quantizing (mostly around the fp16 block weights per 32 params), so they’re able to pack more in to just so few GB.
The tradeoff with unsloth seems to be that it’s more directly applicable for other off the shelf open models, whereas with Bonsai you’d have to wait for them to release another one
This is a great area of research with many players. See also ByteShape https://byteshape.com/blogs/Qwen3.8-27B/ (I'm not affiliated)
There are so many of us with weak GPUs and these updates are massive wins. Still waiting for the 2080 moment where this 27B param model and a reasonable kv cache can exist and function at slow, but overnight-burning speeds ..
This would depend hugely on the price of electricity, however I'm pretty sure where I am (at roughly 25c/kwh in usd) the price of running qwen costs more than 6.1-luna per token. So other than privacy (which doesn't matter for my work) there isn't any justification for running locally. Is privacy purely what people justify local models on?
Doesn't Qwen3.8 do much better than Luna?
As an experiment I took, not this current quant, of Qwen 3.8 27B and had it analyze some of my writing in CPU-only mode with llamacpp. Tried a Q4 and Q5. It was dreadfully slow, but you fire that off before bed, and you can get results the next day. This was a ~6000 word markdown file, took maybe 10 minutes before it started generating anything at all, but then it just cranks away. I would need to tune down the default thinking intensity, but it otherwise did the job.
Not compatible with vibe coding, but if you want a local tool to review content, it is possible today.
I’ve been using Bonsai 2 (also based off of Qwen3.8 27B) on my mac mini in this way (using pizza-bot for the harness). In my case I’ve been using more for deep web search type stuff. It’s been a nice workflow to kick off a request in the evening and come back to a 1 page report on it the next morning.
How does it compare to the 3.8-27B versions from TensorFold? I.e. does it make sense to run unsloth via TensorFold engine?
Figuring out local llms , but it did built me single page html . On 12 GB VRAM and 16 GB ram windows .
Give us more details. What kind of hardware? What kind of setup?
Just installed Qwen 3.8 Splash tonight (doesn’t appear on linked page). Super impressed. Easily averaging 45 tps, even with thinking on. Totally acceptable. Running M4 Mac mini, 48GB, LM Studio.
How do the smallest quants of this compare to Bonsai 2?
Bonsai 2 does some additional tuning in addition to just quantizing (mostly around the fp16 block weights per 32 params), so they’re able to pack more in to just so few GB.
The tradeoff with unsloth seems to be that it’s more directly applicable for other off the shelf open models, whereas with Bonsai you’d have to wait for them to release another one
Bonsai 2 doesn't work at all, Unsloth's smallest... also doesn't work, but is still better.
Bonsai's technology would be better suited for 100s of B models.
Awesome! I wonder how it does at Q5+ compared with v3.0