CredMaster Releases
49 releases of knavesec/CredMaster
- v0.7.10
A bug fix release addressing issues where hotpatching failed when the bin target's crate name differed from the package name.
Jul 30, 2026
- v0.26.0
A major release featuring new Inkling model family support, DeepSeek-V4 performance optimizations, and expanded hardware support via ROCm and XPU.
Jul 27, 2026
- 43.281.0
A minor feature release introducing the ability to automatically update static data.
Jul 26, 2026
- 43.280.5
A maintenance release focusing on bug fixes, dependency updates, and code refactoring for the worker repository logic.
Jul 25, 2026
- 43.280.4
This release focuses on dependency maintenance and security updates, upgrading the pnpm package manager and addressing a security vulnerability in the tar library.
Jul 25, 2026
- 43.280.3
The update includes a bug fix for the base image and introduces internal refactoring to improve release age and confidence checks within the repository worker.
Jul 24, 2026
- 43.280.2
This release corrects a logic error in the preset helper function that pinning GitHub action digests to semantic versioning.
Jul 24, 2026
- runtime-llamacpp-v0.1.9FunASR llama.cpp runtime v0.1.9
This release updates the FunASR llama.cpp runtime with a critical security patch for YAML loading and introduces new hardware acceleration options.
Jul 24, 2026
- v0.25.1
A patch release containing two targeted bug fixes: one to prevent blocking startup when system FFmpeg is missing, and another to prevent garbage output in mixed-dtype quantization fusions.
Jul 14, 2026
- v0.25.0
A major release making Model Runner V2 the default execution path, removing legacy PagedAttention, and significantly improving the Transformers backend performance while adding support for many new models and advanced features.
Jul 11, 2026
- v0.24.1
This patch release addresses a performance regression in the Qwen-Image model and restores the default operator priority for RMSNorm to ensure correct model execution.
Jul 10, 2026
- v0.24.0
This release features 285 commits from 112 contributors, introducing major improvements to the Omni runtime, expanded multimodal model support for TTS and diffusion, and rebases with vLLM 0.23 and 0.24.
Jul 6, 2026
- runtime-llamacpp-v0.1.4FunASR llama.cpp runtime v0.1.4
Runtime update adding AVX2 support for faster execution and fixing SIGILL crashes.
Jun 29, 2026
- runtime-llamacpp-v0.1.3FunASR llama.cpp runtime v0.1.3
Rebuild of the runtime disabling high-ISA CPU flags to fix SIGILL crashes.
Jun 28, 2026
- v4.4.5v4.4.5 — Version Bump for Republish
A version bump for republishing purposes without code changes.
Jun 22, 2026
- v4.4.4v4.4.4 — Dependency Updates
Maintenance release focused on updating dependencies including the web framework hono and build tools.
Jun 22, 2026
- runtime-llamacpp-v0.1.2FunASR llama.cpp runtime v0.1.2
Self-contained binary runtime for SenseVoice ASR with new quantized GGUF models.
Jun 21, 2026
- runtime-llamacpp-v0.1.1FunASR llama.cpp runtime v0.1.1
This release provides a standalone runtime for FunASR llama.cpp, enabling local execution of SenseVoice and Paraformer speech recognition models without requiring Python or PyTorch.
Jun 21, 2026
- docker/egress/v1.1.1components/egress 1.1.1
This release introduces configurable TLS transport for the credential vault and eliminates DNS race conditions during crash restarts. It also adds support for credential vault bindings with allow policies and updates documentation.
Jun 16, 2026
- server/v0.2.0server 0.2.0
This major version introduces a new Credential Vault system for secure secret injection and flattens the execd install path structure. It also fixes several bugs related to proxy handling, conflicting endpoint modes, and pool-mode CRD crashes.
Jun 15, 2026