中文EN

pingud98/llama_cpp_for_radxa_dragon_wing_q6a

memorygitMIT untested

Modifiers to get Llama.cpp working using the built in NPU on the Radxa Dragon Wing Q6A SBC (Qualcomm QCS6490 cpu). Hacked together with Claude Code and Deepseek V4 Flash. It works, but the overall performance for TG is poor, ingestion is super fast - but overall it's a disappointment. I'm now trying

🔌 Network
undeclared
⏳ Sandbox
untested

Who this is for

  • anyone who wants the agent to remember long-running context
下载量 / Downloads
0

Install

dsh plugin --profile web add pingud98/llama_cpp_for_radxa_dragon_wing_q6a

Sandbox test

This plugin has not been tested yet. Submit a test →

Links

GitHub · npm

Frequently asked

How do I install pingud98/llama_cpp_for_radxa_dragon_wing_q6a?
Run dsh plugin --profile web add pingud98/llama_cpp_for_radxa_dragon_wing_q6a in DSH, or paste the package name into the DSH Desktop plugin page.
What if it fails to install?
Check your DSH version first (dsh --version), then look at the first error in the test-results section. If it still fails, open an issue on the repository.
How do I uninstall it?
Run dsh plugin --profile web remove pingud98/llama_cpp_for_radxa_dragon_wing_q6a.

Same category

Compare with peers

PluginDownloadsVerdictSize
dsh-mnemon83.8Kpass17.48 MB
@vectorize-io/hindsight-coding-agents80.4Kpass16.51 MB
dsh-soul-md41.2Kpass11.79 MB
@openviking/dsh-memory-plugin39.4Kpass19.9 MB
dsh-prime-memory30.4Kinstall_failed—
dsh-plugin-memory26.1Kpass7.75 MB
meow-memory25.8Kpass1.78 MB
@restregular/dsh-memory16.7Kpass1.32 MB
dsh-mnemon↓ 83.8K
@vectorize-io/hindsight-coding-agents↓ 80.4K
dsh-soul-md↓ 41.2K
@openviking/dsh-memory-plugin↓ 39.4K
dsh-prime-memory↓ 30.4K
dsh-plugin-memory↓ 26.1K