Search results for Qwen・
Qwen・ community
One keyword maps to one global community path.
People
Not Found
Tweets including Qwen・
Qwen 3.8 Max 0902 is now live in Command Code.
It's an upgraded Qwen 3.8 Max:
· 2.4T params
· 1M context
· Post-trained on coding & cowork
Available on GOAT, Pro, Max and API
Qwen 3.8 27B is still being seriously underrated
i've been running it locally on my RTX 3090 with 24GB VRAM, and it's insane how close it gets to opus 4.8 in some areas
not in raw frontend/design output -- but where qwen gets really interesting is reasoning efficiency
Show more
qwen 3.8 27 B should've shaken openai and anthropic
we're talking about a 27B open-weight model you can run locally on a machine with roughly 24–32 GB of memory, depending on the quantization
tied with gpt-5.6 luna + gemini 3.6 flash, and 1 point behind 5.6 terra
insane!
Show more
Qwen 3.8-Next released with a great tech report with a ton of experimental results specifically surrounding architecture and pretraining. Pleasantly surprising
Paper thread:
Qwen 3.8 Flash is live on AI Gateway.
• 𝚊𝚕𝚒𝚋𝚊𝚋𝚊/𝚚𝚠𝚎𝚗𝟹.𝟾-𝚏𝚕𝚊𝚜𝚑
• Text and image input, 1M context window
Qwen 3.8 Flash-Next is alive on MLX-Serve, under 128gb, working on performance next. MTP gets ~20tok a second, no MTP gets 30+ ... investigating why..
Qwen Flash Next is out, and not really 3.8, it's 4.0.. trying to add support for it now in MLX-Serve, will have a preview build later today. Im on Starlink, so download/upload is a bit slow.
Expected ram needed @ 4bit: ~80gb.
Show more
Qwen 3.8 27B is surprisingly usable on an M4 Pro MacBook Pro with 24GB RAM.
Qwen 3.8 27B uncensored is actually dangerous 🤯