Qwen 3.6 MoE 8bit base request and Roadmap past preview
Hello 🤗, I love your work! It's a treat for us all.
I want to ask some questions, hopefully you won't mind.
When might it go past the preview phase?
Can we have Edge0 but with Qwen 3.6 35b a3b 8bit as base?
Please please 🥺
One correction on our side: edge0-35b is built on Qwen3.6-35B-A3B — we had it labeled "Qwen3.5-MoE" in the README and model card by mistake. It's fixed now.
On preview: no fixed date yet, we're still working on the agentic side.
For an 8-bit 35B, honestly not in the short term. But we're building the ecosystem so new models and precisions can be adapted quickly. Tell us what you'd like and we'll see how fast we can get there.
One correction on our side: edge0-35b is built on Qwen3.6-35B-A3B — we had it labeled "Qwen3.5-MoE" in the README and model card by mistake. It's fixed now.
On preview: no fixed date yet, we're still working on the agentic side.
For an 8-bit 35B, honestly not in the short term. But we're building the ecosystem so new models and precisions can be adapted quickly. Tell us what you'd like and we'll see how fast we can get there.
Qwen3.6? Myan I need to try it. I'm so hopeful I might cry 😭, I see the future of my device in this project, pretty sure a lot does. MoE is the future! Thanks Edge0!!!