ParaVT-8B

by ParaVT

Agentic long-video model trained for parallel temporal tool calls

544dl/month

8B classparams

HuggingFace Use in Pipeline

Identifiers

Model ID

ParaVT/ParaVT-8B

Feature URI

mixpeek://video_extractor@v1/paravt_8b_v1

Overview

ParaVT-8B is a video-text-to-text model focused on long-video understanding through tool use. Its model card describes a parallel video tool-calling approach where the model can dispatch multiple temporal crop requests in one turn instead of walking sequentially through a video.

On Mixpeek, ParaVT is relevant when an agent needs to search long clips and then decide which time windows to inspect. It belongs after retrieval: use vector, transcript, or scene search to narrow the corpus, then let ParaVT reason over candidate spans and request focused temporal crops.

Architecture

Final post-RL ParaVT checkpoint based on Qwen3VLForConditionalGeneration and Qwen/Qwen3-VL-8B-Instruct. The release uses cold-start SFT followed by PARA-GRPO reinforcement learning for parseable, parallel temporal tool-calling behavior.

Mixpeek SDK Integration

import { Mixpeek } from "mixpeek";

const mx = new Mixpeek({ apiKey: "API_KEY" });

await mx.collections.ingest({
  collection_id: "long-video-agent",
  source: { url: "s3://field-video/day-shift.mp4" },
  feature_extractors: [
    { feature: "transcription", model: "openai/whisper-large-v3" },
    { feature: "visual_embeddings", model: "facebook/vjepa2-vitl-fpc64-256" },
    { feature: "scene_caption", model: "ParaVT/ParaVT-8B" }
  ]
});