A model is still running

A model is still running or downloading. Leaving will stop the current operation. Stay on this page to let it finish.

Skip to content
LocalAI.toolsBeta

Join the waitlist

Get an email when the new apps and model runs launch.

No account needed.

SigLIP Base Patch16 224 — Zero-Shot Image Classification

Compare an image with your own English category descriptions. Returns a matching score for each label; scores are independent and are not a guaranteed classification of everything in the image.

Browser inferencePowered by WebAI · Private · Unlimited runs
About this model

Compare an image with your own English category descriptions. Returns a matching score for each label; scores are independent and are not a guaranteed classification of everything in the image.

Created by
Google
License
Apache-2.0
Model released
Unknown
Source repository created
Sep 30, 2023
Model weights · Q4
223.9 MB

Best for

  • Zero-shot image categorization with custom labels.
  • Image-text matching and exploratory content organization.

Keep in mind

  • Scores are independent sigmoid probabilities and do not sum to one.
  • Accuracy depends heavily on label wording, image framing, resolution, and visual domain.
  • The model was trained on English image-text pairs; non-English labels are not validated by this worker.
  • Predictions can reflect training-data bias and must not be used as authoritative safety, identity, medical, or legal judgments.

Processed on your deviceInference runs in this browser. Input/output sharing requires consent for each run. Basic run logs include device information and approximate connection location. Model and runtime files download from external hosts. Saved downloads may be cleared by your browser.

Sources & technical details
Original model
google/siglip-base-patch16-224
Release verification
Unknown
Release evidence
Exact original checkpoint public release date has not been verified. Repository creation, training, paper, artifact conversion and update dates are not release evidence.
Source updated
Sep 26, 2024
Browser model files
onnx-community/siglip-base-patch16-224-ONNX
Files updated
Sep 8, 2026
Artifact revision
9f0328bdd0eb4f62d0f6c10625f43d882b32102b

Input

IMAGE

Sample result

SigLIP Base Patch16 224 — Zero-Shot Image Classification
Advanced settings 1
English template containing one {} placeholder. Change the wording to suit your labels while retaining the required {} placeholder. Different wording changes scores and rankings; longer or more elaborate prompts are not necessarily better.

Output

Your result will appear here

Add input or try the sample, then run the model.

Benchmark evidence

Saved worker evidence is separate from benchmarks run on your device. It does not guarantee speed or quality on every browser.

SigLIP Base Patch16 224 — Zero-Shot Image Classification

12 configurations tested
Passed10
Failed2
TestedSep 11, 2026
RuntimeTransformers.js 4.2.0

Quality criteria

Finite scores in [0,1], unique labels, expected person label ranked first, and its score is at least 10 times the strongest negative.

PrecisionDeviceResult
Q4GPU · WebGPUAcceleratedPassedFast
UINT8GPU · WebGPUAcceleratedPassedModerate
INT8GPU · WebGPUAcceleratedPassedModerate
BNB4GPU · WebGPUAcceleratedPassedModerate
Q8GPU · WebGPUAcceleratedPassedSlow
UINT8CPU · WASMPassedModerate
INT8CPU · WASMPassedModerate
Q8CPU · WASMPassedModerate
BNB4CPU · WASMPassedModerate
Q4CPU · WASMPassedModerate
FP32CPU · WASMFailed—
FP32GPU · WebGPUFailed—
Test setup & sample details
Quality gate
Finite scores in [0,1], unique labels, expected person label ranked first, and its score is at least 10 times the strongest negative.
Tested
2026-09-11T10:36:59.461Z
Environment
Not recorded
Runtime
Transformers.js 4.2.0
Artifact revision
9f0328bdd0eb4f62d0f6c10625f43d882b32102b
Fixture
{
  "file": "modnet-portrait-pexels-5965592.jpg",
  "url": "https://f005.backblazeb2.com/file/hubters-webai-public/assets/benchmarks/modnet-portrait-pexels-5965592.jpg",
  "size": 78124,
  "sha1": "38212739a4a4f0be25d84863cf991d7fb6c6d818",
  "width": 1024,
  "height": 683,
  "license": "Pexels License",
  "labels": {
    "file": "siglip-base-patch16-224.json",
    "url": "https://f005.backblazeb2.com/file/hubters-webai-public/assets/benchmarks/siglip-base-patch16-224.json",
    "size": 460,
    "sha1": "fecc3379e545c31c5fc0dbcd8ca8777189f661e1"
  }
}
Rerun when
artifact or tokenizer revision change; runtime version change; fixture, labels, or quality-gate change; browser/backend behavior change
Per-case quality & failure details
[
  {
    "precision": "fp32",
    "device": "wasm",
    "status": "fail",
    "quality": "Not embedded",
    "runtime": "Not embedded",
    "failure": "cold load exceeded the bounded 15-minute case timeout"
  },
  {
    "precision": "fp32",
    "device": "webgpu",
    "status": "fail",
    "quality": "Not embedded",
    "runtime": "Not embedded",
    "failure": "network error during the 813 MB cold load"
  },
  {
    "precision": "q8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "topToNegativeRatio": 409.1445858649567
    },
    "runtime": {
      "loadMs": 114515.10000002384,
      "inferenceMs": 1586.5,
      "totalMs": 117637.90000009537,
      "imagesPerSecond": 0.6303183107469272,
      "grade": "very-slow"
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "topToNegativeRatio": 466.89075209017426
    },
    "runtime": {
      "loadMs": 135540.5,
      "inferenceMs": 2057.899999976158,
      "totalMs": 138845.60000002384,
      "imagesPerSecond": 0.4859322610484404,
      "grade": "very-slow"
    }
  },
  {
    "precision": "int8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "topToNegativeRatio": 409.1445858649567
    },
    "runtime": {
      "loadMs": 138866.89999997616,
      "inferenceMs": 1514.5,
      "totalMs": 141686,
      "imagesPerSecond": 0.6602839220864972,
      "grade": "very-slow"
    }
  },
  {
    "precision": "int8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "topToNegativeRatio": 466.89075209017426
    },
    "runtime": {
      "loadMs": 22863.099999904633,
      "inferenceMs": 1630.8000000715256,
      "totalMs": 26299,
      "imagesPerSecond": 0.6131959774074938,
      "grade": "very-slow"
    }
  },
  {
    "precision": "uint8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "topToNegativeRatio": 304.0296291817937
    },
    "runtime": {
      "loadMs": 164437.70000004768,
      "inferenceMs": 1491.0999999046326,
      "totalMs": 167500.5,
      "imagesPerSecond": 0.6706458319790475,
      "grade": "very-slow"
    }
  },
  {
    "precision": "uint8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "topToNegativeRatio": 501.96357817718894
    },
    "runtime": {
      "loadMs": 137726.5,
      "inferenceMs": 1606.8999999761581,
      "totalMs": 140654.69999992847,
      "imagesPerSecond": 0.6223162611331365,
      "grade": "very-slow"
    }
  },
  {
    "precision": "q4",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "topToNegativeRatio": 163.48716583218723
    },
    "runtime": {
      "loadMs": 58338.199999928474,
      "inferenceMs": 1794.8000000715256,
      "totalMs": 61489.39999997616,
      "imagesPerSecond": 0.5571651437264031,
      "grade": "very-slow"
    }
  },
  {
    "precision": "q4",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "topToNegativeRatio": 163.48748126862228
    },
    "runtime": {
      "loadMs": 67780.40000009537,
      "inferenceMs": 277.7999999523163,
      "totalMs": 69340.70000004768,
      "imagesPerSecond": 3.599712023656039,
      "grade": "good"
    }
  },
  {
    "precision": "bnb4",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "topToNegativeRatio": 305.14637045620634
    },
    "runtime": {
      "loadMs": 30792.700000047684,
      "inferenceMs": 1597.8999999761581,
      "totalMs": 33703.90000009537,
      "imagesPerSecond": 0.6258213905844676,
      "grade": "very-slow"
    }
  },
  {
    "precision": "bnb4",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "topToNegativeRatio": 305.1469666895808
    },
    "runtime": {
      "loadMs": 84735,
      "inferenceMs": 1652.6000000238419,
      "totalMs": 87679.89999997616,
      "imagesPerSecond": 0.6051071039486706,
      "grade": "very-slow"
    }
  }
]

Something not quite right?

Tell us what happened with SigLIP 1. Loading, results, or the page itself—we’re here for all of it.

Attachments Optional

Up to 3 files · 50 MB each

Browser, OS, CPU/GPU, memory, screen, storage, and network details are included automatically. Model inputs, results, and browsing history are not collected.