Skip to content
Navigation Menu
Sign in
Products
Platform
GitHub
Repository hosting, code review, and collaboration
Copilot
AI coding agent for tasks, projects, and automated work
Build & Ship
Actions
CI/CD and automation
Code Quality
Quality analysis, fixes, and enforcement
Codespaces
Cloud development environments
Secure
Code Security
Vulnerability detection and remediation
Secret Protection
Secret scanning and leak prevention
All Features
Developers
Learn
Documentation
Training
Certifications
Build
Build with Copilot
Customize Copilot
Integrations
Explore
Trending repos
GitHub Sponsors
Developer articles
Changelog
Changelog
Your source for the latest features, improvements, and fixes across GitHub products, all in one place.
Businesses
By Stage
Enterprises
Team
Startups
Nonprofits
By Industry
Healthcare
Financial services
Manufacturing
Government
See all industries
Evaluate GitHub
Why GitHub
Customer Stories
Business insights
Trust Center
Add-ons
GitHub Advanced Security
Enterprise-grade code security
Copilot for business
Enterprise-grade AI features
Premium Support
Enterprise-grade 24/7 support
Resources
Learn & Connect
Documentation
Resource center
Blog
Events & webinars
Community Forum
Customer Stories
Integrations
Support
Customer support
Training
GitHub Status
Company
About
Partners
Changelog
GitHub Universe
GitHub Universe
The event for developers, by developers — talks, workshops, and what’s next for GitHub.
Pricing
Search
/
Sign in
Sign up
Appearance settings
You signed in with another tab or window.
Reload
to refresh your session.
You signed out in another tab or window.
Reload
to refresh your session.
You switched accounts on another tab or window.
Reload
to refresh your session.
Dismiss alert
{{ message }}
Tractables
pyjuice
Repository navigation
Code
Issues
3
(3)
Pull requests
Actions
Projects
Security and quality
Insights
More
items
Pc perf followups
- #58
#58
Merged
liuanji
merged 15 commits into
main
Tractables/pyjuice:main
from
pc-perf-followups
Tractables/pyjuice:pc-perf-followups
Copy head branch name to clipboard
Oct 8, 2026
Conversation
Commits
15
(15)
Checks
Files changed
Merged
Pc perf followups
#58
liuanji
merged 15 commits into
main
Tractables/pyjuice:main
from
pc-perf-followups
Tractables/pyjuice:pc-perf-followups
Copy head branch name to clipboard
Commits
Commits on Oct 6, 2026
fix(tdp): key the top-down CUDA-graph cache by buffer memory, not id()
Show description for 8d60320
liuanji
committed
8d60320
View commit details
Copy full SHA for 8d60320
Browse repository at this point
fix(cudagraph): the recording step was wrong, and graphs replayed under other options
Show description for 5fd39c5
liuanji
committed
5fd39c5
View commit details
Copy full SHA for 5fd39c5
Browse repository at this point
perf(tensorcircuit): keep buffer addresses across batch sizes, so CUDA graphs are reused
Show description for 47c943d
liuanji
committed
47c943d
View commit details
Copy full SHA for 47c943d
Browse repository at this point
perf(softevi): place the dense index's references in one kernel
Show description for abce8bf
liuanji
committed
abce8bf
View commit details
Copy full SHA for abce8bf
Browse repository at this point
Commits on Oct 7, 2026
perf(sum-layer): offer the element-flow autotuner a second tile family
Show description for 304e1c6
liuanji
committed
304e1c6
View commit details
Copy full SHA for 304e1c6
Browse repository at this point
perf(autotune): drop a much slower candidate after one timed run
Show description for af23745
liuanji
committed
af23745
View commit details
Copy full SHA for af23745
Browse repository at this point
test(softevi): run the dense top-k path on HCLT, PD and RAT-SPN
Show description for 640218d
liuanji
committed
640218d
View commit details
Copy full SHA for 640218d
Browse repository at this point
fix(lowrank): hand the backward the state of the forward it follows
Show description for 461eeb1
liuanji
committed
461eeb1
View commit details
Copy full SHA for 461eeb1
Browse repository at this point
perf(blockscale): key the per-row gate-stride cache without reading the device
Show description for beed338
liuanji
committed
beed338
View commit details
Copy full SHA for beed338
Browse repository at this point
fix(tensorcircuit): drop recorded graphs when partial evaluation changes or the circuit moves
Show description for 40457fc
liuanji
committed
40457fc
View commit details
Copy full SHA for 40457fc
Browse repository at this point
feat(cuda-graphs): replay a recorded pass only under what it was recorded with; opt-in compile flag
Show description for 5d96b50
liuanji
committed
5d96b50
View commit details
Copy full SHA for 5d96b50
Browse repository at this point
perf(layers): host-side cdiv and next_power_of_2 in the sum and product layers
Show description for 7d67325
liuanji
committed
7d67325
View commit details
Copy full SHA for 7d67325
Browse repository at this point
fix(tensorcircuit): release buffer storage after a much smaller batch
Show description for 1751341
liuanji
committed
1751341
View commit details
Copy full SHA for 1751341
Browse repository at this point
perf(layers): cut per-call host overhead in the sum/product layers and the kernel launcher
Show description for 179dc80
liuanji
committed
179dc80
View commit details
Copy full SHA for 179dc80
Browse repository at this point
Commits on Oct 8, 2026
test(distributions): seed test_discrete_logistic_nodes
Show description for 36b330c
liuanji
committed
36b330c
View commit details
Copy full SHA for 36b330c
Browse repository at this point
You can’t perform that action at this time.
