We've gotten some great medium sized models lately (DSV4 Flash 0731, Inkling Small, Laguna S 2.1, Step 3.7 Flash) but does anybody else want to see some new 70-80b contenders?
英文标题：We've gotten some great medium sized models lately (DSV4 Flash 0731, Inkling Small, Laguna S 2.1, Step 3.7 Flash) but does anybody else want to see some new 70-80b contenders?
I can run the mediums, but sometimes I want a faster option that's smarter than Qwen 27B/35B. On my hardware I get like 500 to 800 tok/s prefill and 16 to 22 tok/s gen on ~120B class models, which is not the worst but it does get a bit annoying on agentic coding tasks. If we could get some new MoE 7
中文
I can run the mediums, but sometimes I want a faster option that's smarter than Qwen 27B/35B. On my hardware I get like 500 to 800 tok/s prefill and 16 to 22 tok/s gen on ~120B class models, which is not the worst but it does get a bit annoying on agentic coding tasks. If we could get some new MoE 7
中文全文翻译正在补齐，可先查看下方英文原文。
English / 原文
I can run the mediums, but sometimes I want a faster option that's smarter than Qwen 27B/35B. On my hardware I get like 500 to 800 tok/s prefill and 16 to 22 tok/s gen on ~120B class models, which is not the worst but it does get a bit annoying on agentic coding tasks. If we could get some new MoE 7