ViperEkura
/
video-promo
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
17 Commits 1 Branch 0 Tags 240 KiB
Python 100%
Go to file
ViperEkura 57abefa47f fix: shift GQA layout down 0.4 to avoid title-input overlap 2026-05-07 14:51:31 +08:00
.gitignore chore: switch .gitignore to whitelist mechanism (only .py, .md, .gitignore) 2026-05-07 14:14:15 +08:00
PROMO_GUIDE.md add project source files 2026-05-06 21:16:57 +08:00
README.md first commit 2026-05-06 21:15:18 +08:00
architecture.py refactor: architecture — boxes left, descriptions right, 4-layer layout 2026-05-07 12:18:33 +08:00
continuous_batching.py refactor: Gantt chart — same 5 requests both sides, wait+run bars, batch boxes connect GPU↔reqs 2026-05-07 00:33:42 +08:00
prefix_cache.py add project source files 2026-05-06 21:16:57 +08:00
render_all.py fix: architecture layout and rewrite continuous_batching with dynamic pipeline animation 2026-05-06 21:48:18 +08:00
transformer.py fix: shift GQA layout down 0.4 to avoid title-input overlap 2026-05-07 14:51:31 +08:00

README.md

AstrAI Promo — 动画渲染

Manim 驱动的技术动画，用于 AstrAI 宣传视频。

依赖

pip install manim
# 或: conda install -c conda-forge manim

渲染单个场景

# 1080p（推荐）
manim -qh promo/continuous_batching.py ContinuousBatching

# 480p 草稿
manim -ql promo/continuous_batching.py ContinuousBatching

# 4K
manim -4k promo/continuous_batching.py ContinuousBatching

渲染全部

python promo/render_all.py

输出到 promo/output/videos/。

场景清单

文件 Scene class 内容 建议时长
transformer.py Transformer GQA 注意力机制详解 (Q/K/V→RoPE→Attention→O) + 规格卡 ~25s
continuous_batching.py ContinuousBatching 4 阶段流水线 + 吞吐对比 ~30s
prefix_cache.py PrefixCache Radix Tree 生长 + 前缀复用 ~30s
architecture.py Architecture 全栈架构逐层展开 ~30s

导入视频剪辑

输出的 .mp4 直接导入 DaVinci Resolve / Camtasia / Premiere Pro。