It's on baby!!! when my 60k openAi credits are done, it'll only be a 20 clinician plan. everything else will be thru qwen's token plan and local inference.
the other frontier models are great and have their place, but I don,t believe I need them majority of the time. These secondary models will do perfectly well 90% of the time at least based on the last month of testing
📢 Meet Qwen4-Max, your new AI teammate for work, creativity, and discovery!
Qwen4-Max is built on a new MoE architecture with native vision and a 1M-token context window.
Highlights:
- Outperforms Qwen3.8-Max with up to ~50% fewer tokens, while average cost per task drops over 60% in our internal evaluation.
- Advances real-world productivity through greater proficiency with professional tools, stronger long-horizon execution, and higher-quality deliverables.
- Handles real tasks, including 3D modeling, game development, manufacturing, finance, law, healthcare, and frontier scientific discovery.
- $1.5/M tokens for input, $4.5/M tokens for output, and $0.075/M tokens for implicit caching — all lower than Qwen3.8-Max, with implicit caching down by 70%.
The API is live on QwenCloud. You can also try Qwen4-Max on Qwen Studio, QwenWork and Qoder.
Start building today! 🚀
- Blog: qwen.ai/blog?id=qwen4-max
- API: qwencloud.com/models/qwen4-m…
- Token Plan: qwencloud.com/pricing/token-…
More to come! The open weights of smaller Qwen4 models will be released. Stay tuned. 👀
99
Jared Gibb, BCBA retweeted
📢 Meet Qwen4-Max, your new AI teammate for work, creativity, and discovery!
Qwen4-Max is built on a new MoE architecture with native vision and a 1M-token context window.
Highlights:
- Outperforms Qwen3.8-Max with up to ~50% fewer tokens, while average cost per task drops over 60% in our internal evaluation.
- Advances real-world productivity through greater proficiency with professional tools, stronger long-horizon execution, and higher-quality deliverables.
- Handles real tasks, including 3D modeling, game development, manufacturing, finance, law, healthcare, and frontier scientific discovery.
- $1.5/M tokens for input, $4.5/M tokens for output, and $0.075/M tokens for implicit caching — all lower than Qwen3.8-Max, with implicit caching down by 70%.
The API is live on QwenCloud. You can also try Qwen4-Max on Qwen Studio, QwenWork and Qoder.
Start building today! 🚀
- Blog: qwen.ai/blog?id=qwen4-max
- API: qwencloud.com/models/qwen4-m…
- Token Plan: qwencloud.com/pricing/token-…
More to come! The open weights of smaller Qwen4 models will be released. Stay tuned. 👀
watching qwen 3.8 27b thought process is a lot like talking to my own spaztic self sometimes. But boy howdy, when it's doing something it's good at, it's good
1
101
running qwen 3.8 27b at home at 35tps feels great.
tiny-tank.pages.dev/
can't wait to see what fun crap i can create when i let a flagship orchestrate alongside me and i let qwen do the heavy lifting.
a brains vs brawn battle
1
87
ok @Alibaba_Qwen I see you and your 1 shot magic ability with qwen 4
May i present to you all: Flappy Turd
flappy-turd.codelessconsulta…
1
2
185
Don’t be a dumbass
npr.org/2026/11/03/nx-s1-602…
1
68
Astra Directing
today im evaluating using Mistral Large 4 and Qwen 3.8 flash next as my implemnters.
Plan
build
feedback
build
Cycle
118
Anyone else out here tried Mistral?
Signed up today to try it out for a month while i setup my local rig. Mistral 4 Le Chonk is a pretty solid competitor for what im planning to do with it.
Looking forward to Qwen 4 dropping in the next few weeks
20/month for GPT with 60k credits banked
14/month for mistral
electricity cost only for Unlimited Local Tokens
2
180
thank you @thsottiaux for skipping resets so that i could know the only important thing.
HOW MUSH I GOT LEFFFFF??????
1
74
Excited to build this rig and to stop paying per token and instead for my own KwH
4x 32GB Crucial DDR4-3200 CL22 (128GB System Ram)
2x 24GB Intel Arc B60 (48GB VRAM total)
AMD Ryzen 7 5700X 3.8ghz 8-core
1TB ssd
She gonna bang out some tokens for me
1
182
Terra 6.1 can eat poop. it's dumb AF compared to Sol and especially against Astra.
Better at using computers, my butt!
88









