Qwen3.8 27B

And the numbers are great!

Not really. I have Qwen3.8 27b building all night long in an agentic loop. Try doing that with your $20 subscription. And once you get into the $200 plans, now you are talking merely months, not years to recover your investment.

I think I got it running at 17-18. Still more than 16 though. And to clarify, that's with kv cache on RAM, not offloaded.

Nope, not worth it. Also, FreeToken targets MoE, and Qwen3.8:27B is a dense model, so you won't get much out of it.

performs well for me

I’ve been struggling to make even a simple 3D car game with AI and Godot. I’ve tried ChatGPT, Qwen 3.8 27B, DS Flash, Muse 1.3... and the results have been pretty bad.

So far with 3.8 i just need to check the end result...

Didn't expect the 27B to be that far behind

q3 xl unsloth. 150k context. hatch 512 and kv q4. Its solid. review luna and sol plans and catch gaps confirmed later by them

it can follow my coding conventions and even document the architecture the right way without much hand holding. No other model that fit on my 16GB VRAM does as well.

You can actually tell the difference between q3 and q4 quants.

Qwen3.8 27B have very similar behavior to Minimax M3 (thinks a lot, check everything, and like to continue working on its own for long session until it's done). It's like a slightly worse M3.

Qwen 3.8 27B is the real deal BUT remember to use froggeric template and/or medium reasoning

it thinks forever and sometimes in a loop

It can hallucinate with historical facts so dojlnt trust too much

qwen3.8 27b is the same workflow but sloooooweeeeeerrrrrrr

it had hallucinated a key detail

It feels like it can never figure out how to update memory

Completely unusable last time I tried it. You got hit with rate limits after the first few minutes of using it. It's fast but cannot sustain its claimed speed before it immediately hits its rate limit. What good is it if you can't finish a

il a réfléchi 15 min et codé 2 min

i see the memory use growing during tool use, with qwen3.8 27B it certainly plays a role

falls apart on multi-step agentic tasks. Tried it on a refactor that needed search plus edit across files and by the third tool call it was citing functions it had already overwritten.