title, date, draft, tags, categories, summary, author
| title |
date |
draft |
tags |
categories |
summary |
author |
| [myai] feat(myai): add POST /generate via StarCoder2-3B (GPU) |
2026-08-11T17:30:58+02:00 |
false |
|
|
Commit a8a32ef par Betty dans myai |
Betty |
Commit a8a32ef
|
|
| Repository |
myai |
| Branch |
main |
| Author |
Betty |
| Hash |
a8a32ef01bfe78e041b3ea636d354300dc751063 |
Description
- New app/codegen.py: loads bigcode/starcoder2-3b at startup via HuggingFace
- POST /generate endpoint: prompt + language → generated code
- Dockerfile: switch to pytorch/pytorch:2.2.0-cuda12.1-cudnn8-runtime base
- requirements.txt: add transformers<5.0 + accelerate
- stack/myai.yml: GPU placement (nvidia.vgpu), VRAM/CPU limits, hf-cache volume
Co-Authored-By: Claude Sonnet 4.6 noreply@anthropic.com
Files Changed
Diff Summary