Qwen2.5-coder is a game changer for secure, private, and local code development in VSCode 🤩 We've been testing the latest Qwen models running locally in VSCode through ollama and CodeGPT, and the results are incredible I highly recommend downloading and trying out these models directly from #CodeGPT.. you'll be impressed!

@Alibaba_Qwen @ollama Love how open-source models are transforming developer collaboration! #Qwen⁠2.5-coder

@dani_avila7 @Alibaba_Qwen 🚀🚀🚀

@Alibaba_Qwen @ollama @codegptAI Does it automatically build in the backdoor for PRC? 😂

@dani_avila7 @Alibaba_Qwen @codegptAI you can run Ollama inside a faraday's cage (and importing models separately) 🫡

@Alibaba_Qwen @ollama @codegptAI Just hype

@Alibaba_Qwen @ollama @codegptAI its so crazy !!! and then you have github copilot just sitting there with claude, gpt .. waiting to help ..

@Alibaba_Qwen @ollama @codegptAI Nice!

@Alibaba_Qwen @ollama @codegptAI It's cool. Seems to be working, still figuring it out

@Alibaba_Qwen @ollama @codegptAI I only have an 8c system with a 3080ti and 14b is the fastest that I can run.

@Alibaba_Qwen @ollama @codegptAI I only use AI for simple auto completes, last I tried codegpt it wasn't that good comparing to copilot or supermaven, did they change anything? Really love local code dev its much safer, would go a step further to buy a beefy AI server and run ollama at home if it was as good.

I've made quite a few changes to the auto-complete feature. Feel free to try it again with any of the open-source models available. If you're going to use the Qwen2.4-coder model, I recommend using version 3b or later. You can also try the CodeGPT Plus or CodeGPT Plus-Turbo models, which are really good 🙌