返回
RCreddit.com

local vibecoding with Qwen 3.8 27B and Godot

LlamaQwen模型发布端侧推理
时间与来源
发布
09/05 11:07
收录
09/05 20:00
来源类型
开发者社区
档位
社区
信源状态
正常
档位是按信源手工设定的编辑判断,不是逐条打分。
AI 摘要

一位开发者使用 Qwen 3.8 27B 和 Godot 探索了“本地振动编码”。起初,开发者对 Godot 不熟悉,它被下载为一个 .exe 文件。随后,开发者利用类似“pi”的工具与 Qwen 3.8 27B 结合,自主研究该主题。此过程涉及运行一个 llama-server 命令,其中包含 Qwen 模型的特定参数,从而使计算机在初始提示后能以最少的用户输入进行工作。

OK, you guys want to start with local vibecoding but don't know how. In the discussions, you recommended that I use Godot. I downloaded it without knowing what it was. I was expecting a library, but it was an .exe file. So I started pi with Qwen 3.8 27B and told it to explore this topic without my help.

I needed to use four prompts. You can see all of them in the screenshots, so you can reproduce my steps. The whole session used 25% of my context, so you can use just 1/4, which is about 64k.

The game works: you can walk around the dungeons, the llamas are dancing, and there are dynamic lights.

llama-server -m /mnt/models2/Qwen/3.8/27B/Qwen3.8-27B-UD-Q8_K_XL.gguf -mm /mnt/models2/Qwen/3.8/27B/Qwen3.8-27B-mmproj-BF16.gguf --host 0.0.0.0 --jinja -fa on --keep 4096 -b 8192 --parallel 1 --ctx-checkpoints 12 --cache-ram 65536 --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0 --presence-penalty 0 --repeat-penalty 1.0 --spec-type ngram-mod --spec-type draft-mtp --spec-draft-n-max 3 --chat-template-kwargs {"preserve_thinking":true}

You need pi or something similar. This way, you can use just a few prompts and let the computer work without your input.