v1.58.4

1.58.4 (2026-03-12)

Bug Fixes

• scheduler: requeue by global fit check in queueing hint (#603) (7e02969)

v1.58.3

1.58.3 (2026-03-11)

Bug Fixes

• compute component template hash from rendered PodSpec instead of… (#601) (7a9b1a4)

v1.58.2

1.58.2 (2026-03-10)

Bug Fixes

• enable privileged mode for hypervisor container (#599) (2852c10)

v1.58.1

1.58.1 (2026-03-05)

Bug Fixes

• copy api/ before go mod download in node-discovery Dockerfile (#590) (0ce3548)

v1.58.0

1.58.0 (2026-03-04)

Features

• add per-GPU resource metrics tracking and update database schema (#589) (e57a35a)

v2.4.2

2.4.2 (2026-02-26)

Bug Fixes

• slove the problem of go lint failure (#581) (bc9358f)

v2.4.1

2.4.1 (2026-02-26)

Bug Fixes

• solve hypervisor and operator some issues (#568) (3c0ad97), closes #558

API Module v0.1.2

Usage

go get github.com/NexusGPU/tensor-fusion/api@v0.1.2

Changes

• fix: solve hypervisor and operator some issues (#568)

v2.4.0

2.4.0 (2026-02-12)

Features

• make api/v1 independently importable as a Go sub-module (#571) (2f2c6a3)

v2.3.6

2.3.6 (2026-02-12)

Bug Fixes

• enhance GPU allocation logic and QoS level calculation (#570) (55025d4)