← DeepSeek Harness Plugins
DeepSeek Harness PluginManifest Valid

dsh-radeon-cloud-patcher

Integrates AMD Radeon Cloud model provider into DeepSeek Harness with per-model reasoning effort levels.

VisionTerminal & TUIDeveloper Tools
0GitHub Stars0ForksUpdated2026-10-17

Install

$ dsh plugin --profile <profile> add @zhz8888/dsh-radeon-cloud-patcher。

Plugin Overview & Capabilities

AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.

source-grounded
dsh-radeon-cloud-patcher is a DeepSeek Harness plugin that registers AMD Radeon Cloud as a model provider without modifying core Harness files or user profiles. It uses an embedded patch layer delegating transport and reasoning parsing to the built-in dsh-llm-pi-ai adapter, offering per-model reasoning effort levels, image input support for compatible models, and secure API key management.

Key Capabilities

  • Adds AMD Radeon Cloud models to the DSH model picker
  • Supports per-model reasoning effort configuration (e.g., off, low, medium, high, max)
  • Provides vision/image input support for multimodal models such as DeepSeek-V4.1-Flash and MiniCPM-V-4.7
  • Manages credentials securely via DSH credential storage (~/.dsh/.credentials.yaml) or environment variables
  • Zero runtime external dependencies, delegating execution cleanly to built-in dsh-llm-pi-ai

Useful For

  • Accessing AMD Radeon Cloud shared AI models directly inside DeepSeek Harness
  • Fine-tuning reasoning and thinking effort levels on supported DeepSeek, Qwen, and MiniCPM models
  • Processing multimodal image and text prompts via Radeon Cloud's hosted endpoints

Who It Fits

  • DeepSeek Harness users with AMD Radeon Cloud API access
  • Developers seeking granular control over model reasoning effort tiers

Documented Limitations

  • Reasoning tokens and generated output share the same max_tokens budget, potentially yielding empty responses if tokens are exhausted during thinking
  • Upstream AMD gateway ignores the stop option, throwing UNSUPPORTED_OPTION if passed
  • Only covers AMD Radeon Cloud Public Free Model APIs (shared endpoint); dynamic dedicated endpoints are not supported
  • Selecting unsupported reasoning tiers triggers hard API errors (400/422) without automatic fallback

DSH Compatibility

Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.

Not tested yetNo runtime compatibility tests have been published yet.

Security Signals

Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.

Dependency Counts

package.json declares 0 runtime, 0 development, 4 peer, and 0 optional dependencies.

info
Dsh Bundle Declared

package.json declares DSH bundle metadata.

info
License Declared

GitHub reports the repository license as MIT.

info
Package Manifest Available

A root package.json was captured and can be inspected by the registry.

info
Source Available

Public GitHub source metadata is available for this registry snapshot.

info

Source & Registry Notes

Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.

Source repository
zhz8888/dsh-radeon-cloud-patcher
Registry source
GitHub · dsh-plugin topic
Registry classification
Plugin
Source checked
ce25bfc · 2026-10-17

This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.

Repository Activity

GitHub stars
0GitHub stars
Forks
0
Open issues
0
Last commit
2026-10-17
Last release
2026-10-15
For maintainers

Maintaining this plugin?

This listing is generated from public repository data. If you maintain this project, you can review the information and share this listing with your users if you find it useful.

Add to README
Listed on DSHPlugin.app