PD ratio planner for xPyD proxy — recommend optimal Prefill:Decode node allocation
Project description
xPyD-plan
PD ratio planner — recommend optimal Prefill:Decode allocation from real benchmark data.
Install
pip install xpyd-plan
# With HTML reports
pip install "xpyd-plan[report]"
# Development
pip install "xpyd-plan[dev]"
Quick Start
# Find optimal P:D ratio from benchmark results
xpyd-plan analyze --benchmark results.json --sla-ttft 200 --sla-tpot 50
# Compare two benchmark runs
xpyd-plan compare --baseline baseline.json --current current.json
# Generate report
xpyd-plan report --format markdown --benchmark results.json --output report.md
License
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
xpyd_plan-0.1.0.tar.gz
(532.3 kB
view details)
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
xpyd_plan-0.1.0-py3-none-any.whl
(467.5 kB
view details)
File details
Details for the file xpyd_plan-0.1.0.tar.gz.
File metadata
- Download URL: xpyd_plan-0.1.0.tar.gz
- Upload date:
- Size: 532.3 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.13
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
7a1daa8d878aacb8d1e6313ca52de880740fbb75455a3665c1bc0a845c879c7d
|
|
| MD5 |
39274c00a38312d1b48a16000db5a428
|
|
| BLAKE2b-256 |
320a0038900f0b7632b3526e88dc57f32edf65f3a5ae65dc5ab3c8622cef91aa
|
File details
Details for the file xpyd_plan-0.1.0-py3-none-any.whl.
File metadata
- Download URL: xpyd_plan-0.1.0-py3-none-any.whl
- Upload date:
- Size: 467.5 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.13
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
8ee13173c7806feb0195d3c0cb97ea6922b4822687a987c1d77697dfa8b4fdc3
|
|
| MD5 |
0a9ddc467cdb6ca15e4513de641d7f25
|
|
| BLAKE2b-256 |
227e32c9b966759dffdab92a4b5f98cece0d9cd7d3847fcb04d0eabecf05092d
|
