DA3-LARGE-1.1

by depth-anything

Depth Anything V3 Large checkpoint for spatial scene retrieval

35Kdl/month

11likes

411Mparams

HuggingFace Run on your data, free

Identifiers

Model ID

depth-anything/DA3-LARGE-1.1

Feature URI

mixpeek://image_extractor@v1/depth_anything_v3_large_v1

Overview

Depth Anything V3 Large is a newer monocular depth checkpoint for estimating dense depth maps from images. It improves the depth channel available to perception pipelines where spatial layout matters as much as semantic content.

On Mixpeek, Depth Anything V3 Large can enrich images or video frames with depth-derived metadata such as foreground ratio, depth bands, relative distance, and scene layout. Agents can use those signals to find clips by spatial evidence before calling a heavier reasoning model.

Architecture

Depth estimation model from the Depth Anything V3 family. It produces dense per-pixel depth maps from single images and is exposed through the HuggingFace depth-estimation pipeline.

Mixpeek SDK Integration

import { Mixpeek } from "mixpeek";

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

// Managed: create a collection over a bucket; Mixpeek runs this model's extractor
const collection = await mx.collections.create({
  namespace_id: "my-namespace",
  collection_name: "my-collection",
  source: { type: "bucket", bucket_ids: ["bkt_your_bucket"] },
  feature_extractor: {
    feature_extractor_name: "depth_estimation",
    version: "v1",
    parameters: { model_id: "depth-anything/DA3-LARGE-1.1" },
  },
});