DSH Plugins Marketplace

DSH Plugins

Plugins

/

Vision & Multimodal

/

dsh-video-localmodel

d

dsh-video-localmodel

Manifest valid

Local-first video screening: ffmpeg samples frames, local Qwen3-ASR writes timecoded subtitles, a local vision model captions and ranks sampled frames, and a local text model folds the transcript into

hasBundlePatch

dsh-video-localmodel

Local-first video screening for DeepSeek Harness: local ASR writes timecoded subtitles, a local vision model captions sampled frames, a local text model folds the transcript into chapters — and the original candidate frames go to the cloud model, which makes the final call.

中文: README.zh.md

Local inference costs time, not money, so this plugin spends minutes locally to save the leading model dozens of frames. The small models are screening and navigation only: timelines, outlines and captions can be wrong in the details. The exported original frames are the evidence.

Pipeline

L0 deterministic  ffprobe + scene cuts + frame sampling        seconds, no model time
L1 local ASR      Qwen3-ASR-0.6B, silence-aware chunks         ~2.4x realtime on CPU
L2 local vision   qwen3-vl:4b, caption + importance 1-5        ~3 s per frame on GPU
L3 local text     qwen2.5:7b, window summaries -> chapters     ~10 s per window
L4 hand-off       SRT + timeline + native-resolution frames    leading model reads pixels

Every stage writes under <workspace>/.video/<file-key>/, so a rerun resumes. ASR is reused while the plan is unchanged; pass force: true to video_local_scan to re-run it.

Tools

ToolWhat it doesCost
video_local_planDetect ffmpeg/Python/ASR weights/Ollama models, report the exact models and estimateno model time
video_local_scanProbe, extract audio, local ASR with timecodes, scene cuts, coarse frames~5 min per 10 min video (CPU)
video_local_captionLocal VLM captions + importance, resumable across calls~3 s per frame (GPU)
video_local_timelineLocal text model: timeline, chapters, keywords, key-frame ranking~30 s
video_local_exportSRT, timeline Markdown, native key frames, best frames attached as images~15 s
video_local_framesOriginal frames for one time range (max 12)seconds

Model choice is explicit: video_local_plan reports the ASR/vision/text models, and video_local_scan refuses to run without plan_id plus confirm_models: true. Changing models or parameters invalidates the plan.

Install

In DSH, open Plugins → + Add plugin and enter:

github:wth373737/dsh-video-localmodel

Then restart DeepSeek Harness once. Installation downloads nothing: the plugin shells out to the ffmpeg, Python and Ollama you already have, and video_local_plan reports what it found before any model time is spent. See Requirements below.

Requirements

  • ffmpeg + ffprobe on PATH (or configured)
  • Python with torch and qwen_asr, plus a local Qwen3-ASR checkpoint
  • Ollama with a vision model (e.g. qwen3-vl:4b) for the standard mode and a text model (e.g. qwen2.5:7b)

All paths are configurable; detection order is plugin config → <workspace>/.video/settings.json → $DSH_HOME/video-localmodel.json → DSH_VIDEO_* env vars → conventional locations.

Output

<workspace>/.video/<key>/
  plan.json probe.json scenes.json transcript.json captions.json timeline.json timeline.md
  frames/                 coarse 512px frames
  export/report.md timeline.md timeline.json subtitles.srt frames/  (native key frames)

Notes

  • Sentence-level SRT timecodes are approximated by character share inside each ASR chunk; word-level alignment would need Qwen3-ForcedAligner.
  • qwen3-vl is a thinking model: the bridge budgets extra num_predict and keeps only content, so reasoning never leaks into captions.
  • Python children get TEMP/NUMBA_CACHE_DIR inside the workspace, which keeps the pipeline alive under a restricted sandbox.
  • Local files only in this version; URL ingestion is planned.

License

MIT. This project ships no video content and provides no way to bypass a site's terms.

Comments

Loading…

Similar plugins

dsh-video-eyes

Read a video with a text-only model: with ffmpeg installed, sample frames at a 0.1-10 second interval into contact sheets where every frame carries its real timecode, extract the audio track and send

Vision & MultimodalManifest valid

★ 0

dsh plugin --profile web add dsh-video-eyes

by Weilv-D

Streams and inspects local video for agents: adaptive motion/settle canvas sampling, deterministic probe, timestamp and uniform frame extraction, and RoI zoom on 42px-grid-aligned canvases. Requires f

Vision & MultimodalManifest valid

★ 0

Apache-2.0

TypeScript

Nov 9, 2026

dsh plugin --profile web add dsh-pf-stream

Subtitle-first video transcription with SRT export, cancellable job controls, and a faster-whisper large-v3 fallback when subtitles are unavailable.

Tools & CapabilitiesVision & MultimodalManifest valid

★ 0

dsh plugin --profile web add dsh-watch-video

by dundunhan

Give text-only DeepSeek Harness agents video understanding: scene-aware frame sampling + VLM + optional ASR transcript fused into timeline evidence. / 给纯文本模型的视频理解插件(场景感知抽帧 + VLM + 可选语音转录)

Vision & MultimodalManifest valid

★ 112

↓ 479/wk

MIT

JavaScript

Sep 19, 2026

dsh plugin --profile web add dsh-video-lens

by Kagurazaka135

Give a text-only agent video understanding: extract timestamped frame files, sample only on scene changes, export GIF clips, probe metadata, and run a one-shot analyse over the frames. Five tools that

Vision & MultimodalManifest valid

★ 0

JavaScript

Oct 17, 2026

dsh plugin --profile web add dsh-video-frames

Video analysis panel in the DeepSeek Harness sidebar. Paste a YouTube or Bilibili link to fetch its subtitles and generate a summary, chapters, a mind map and citation-backed Q&A over the transcript,

Tools & CapabilitiesManifest valid

★ 1

MIT

JavaScript

dsh plugin --profile web add dsh-video-insight