Commits

History for

Commits on Jul 10, 2026

fix(distributed): authenticate coordinator worker registration
Show description for 35db753
dndungu
committed
35db753
Copy full SHA for 35db753

Commits on Jul 4, 2026

fix(distributed): wire mTLS into worker gRPC server, refuse non-loopback without TLS
Show description for d6b2562
dndungu
committed
d6b2562
Copy full SHA for d6b2562

Commits on Apr 28, 2026

refactor(layout): move health/ to serve/health/ (T124.3.1)
dndungu
committed
712336e
Copy full SHA for 712336e

Commits on Mar 26, 2026

refactor(fsdp): migrate GGUF writer to ztensor/gguf
Show description for 01af3dd
dndungu
committed
01af3dd
Copy full SHA for 01af3dd

Commits on Mar 25, 2026

fix(distributed): add stop channel to coordinator reaper goroutine
Show description for d89de2d
dndungu
committed
d89de2d
Copy full SHA for d89de2d

Commits on Mar 18, 2026

docs(stability): label all sub-packages with stable/beta/alpha maturity
Show description for 2314d39
dndungu
committed
2314d39
Copy full SHA for 2314d39
docs(blog): add 5 launch blog posts
Show description for 16e935e
dndungu
committed
16e935e
Copy full SHA for 16e935e
test(fsdp): add 2-GPU FSDP integration test
Show description for d998760
dndungu
committed
d998760
Copy full SHA for d998760
feat(distributed): add FSDP distributed checkpoint save/load (T10.5)
Show description for 437ff65
dndungu
committed
437ff65
Copy full SHA for 437ff65
feat(distributed): add sharded AdamW optimizer state (T10.4)
dndungu
committed
6984c2a
Copy full SHA for 6984c2a
feat(distributed): add FSDP gradient accumulation (T10.3)
dndungu
committed
0d04858
Copy full SHA for 0d04858
feat(distributed): add FSDP sharded module (T10.1)
Show description for f23e510
dndungu
committed
f23e510
Copy full SHA for f23e510
feat(distributed): add NCCL AllGather+ReduceScatter via purego dlopen (T10.2)
Show description for a32fe8d
dndungu
committed
a32fe8d
Copy full SHA for a32fe8d

Commits on Mar 16, 2026

docs(distributed): add package-level documentation
dndungu
committed
409562b
Copy full SHA for 409562b
chore(lint): vet/lint clean after Q5K/Q6K dequant changes
Show description for 071d4d9
dndungu
committed
071d4d9
Copy full SHA for 071d4d9
Fix test failures from import migration and ZMF removal
Show description for 13d61a8
dndungu
committed
13d61a8
Copy full SHA for 13d61a8
Rewrite imports from zerfoo to ztensor/ztoken modules
Show description for c60d3f7
dndungu
committed
c60d3f7
Copy full SHA for c60d3f7

Commits on Mar 4, 2026

fix(distributed): fix NCCL strategy import and logger type errors
Show description for 3133a3b
dndungu
committed
3133a3b
Copy full SHA for 3133a3b

Commits on Mar 3, 2026

feat(distributed): add NcclStrategy for GPU-native gradient exchange
Show description for 2368731
dndungu
committed
2368731
Copy full SHA for 2368731
feat: generic ML framework with GPU engine, model import, and production infra
Show description for c80f934

dndungu
and
claude
committed
c80f934
Copy full SHA for c80f934

Commits on Aug 25, 2025

refactor(distributed/coordinator): streamline coordinator logic
dndungu
committed
45e84f7
Copy full SHA for 45e84f7
refactor(distributed): refine all-reduce implementation
dndungu
committed
5900351
Copy full SHA for 5900351

Commits on Aug 24, 2025

distributed/coordinator: format indentation and keep context-aware listener pattern
dndungu
committed
c3942b9
Copy full SHA for c3942b9
distributed: refactors and test updates for all-reduce and coordinator/network manager
dndungu
committed
cb2b23f
Copy full SHA for cb2b23f

Commits on Aug 22, 2025

refactor: Rename Tensor to TensorNumeric to avoid conflict with Tensor interface
dndungu
committed
48b0366
Copy full SHA for 48b0366

Commits on Aug 4, 2025

Continue systematic lint fixes: reduce issues from 51 to 9
Show description for b234ab1
dndungu
committed
b234ab1
Copy full SHA for b234ab1
Fix lint issues: unused parameters, package comments, and documentation
Show description for 57f67b6
dndungu
committed
57f67b6
Copy full SHA for 57f67b6
fix: Resolve unused-parameter revive issues in distributed/network_manager_test.go
dndungu
committed
a2fa402
Copy full SHA for a2fa402
fix: Resolve unused-parameter revive issues in distributed/custom_mocks_test.go
dndungu
committed
d4f7eac
Copy full SHA for d4f7eac
fix: Resolve exported revive issue in distributed/interfaces.go
dndungu
committed
d6e045a
Copy full SHA for d6e045a
fix: Resolve unused-parameter revive issue in distributed/coordinator/coordinator_test.go
dndungu
committed
0a7b5ec
Copy full SHA for 0a7b5ec
fix: Resolve exported revive issue in distributed/coordinator/coordinator.go
dndungu
committed
c0064a0
Copy full SHA for c0064a0
fix: Resolve typecheck errors in distributed/coordinator/coordinator_test.go by renaming t to subT in t.Run anonymous functions
dndungu
committed
13865c0
Copy full SHA for 13865c0
Revert "fix: Resolve typecheck errors in distributed/coordinator/coordinator_test.go"
Show description for b98dc47
dndungu
committed
b98dc47
Copy full SHA for b98dc47
fix: Resolve typecheck errors in distributed/coordinator/coordinator_test.go
dndungu
committed
8b7edf7
Copy full SHA for 8b7edf7