Native LLM-provider vision bridge: images pasted in the chat are described by a vision model (Qwen3-VL via pi-ai/llama.cpp) and the text description is fed to text-only DeepSeek for the reply — image admission, routing and compaction all run through harness-native mechanisms, with an LRU description cache and 503 retry.
Author: Einskyle · Vision & Multimodal · Catalog snapshot: 2026-09-04Native LLM-provider vision bridge: images pasted in the chat are described by a vision model (Qwen3-VL via pi-ai/llama.cpp) and the text description is fed to text-only DeepSeek for the reply — image admission, routing and compaction all run through harness-native mechanisms, with an LRU description cache and 503 retry.
Imported third-party catalog description; not a Registry verification conclusion.Einskyle/dsh-llm-vision-bridge
4 stars · 234 downloads
Signals were provided by the source catalog and do not imply quality or safety. 2026-09-04No exact-version test is available.
No Registry security conclusion is available.
No functional smoke evidence is available.
Plugin permissions have not been resolved.
Version and integrity have not been resolved.
Describe your task in a DSH conversation and mention dsh-llm-vision-bridge. The plugin manager can find this candidate through Registry, then inspect its actual source and version locally.
dsh-llm-vision-bridgeExact version not resolvedgithub:Einskyle/dsh-llm-vision-bridgeExact version not resolved