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.

Qwen3 Embedding

Compare meaning, explore text vectors, and build semantic search on your device.

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

Convert multilingual text into 1024-number vectors for semantic search and document matching. Use the expected instruction formatting for queries; the model returns vectors, not answers.

License
Apache-2.0
Model released
Unknown
Source repository created
Jun 3, 2025
Model weights · FP16
1.2 GB

Best for

  • Multilingual and cross-lingual semantic search and retrieval.
  • Clustering, similarity, and classification prototypes.

Keep in mind

  • Retrieval quality depends on language, domain, input quality, and correct query instructions.
  • Semantic similarity is not factual verification and can encode training-data bias.
  • Long inputs are truncated and should be chunked.
  • Quantization can change ranking margins.

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
Qwen/Qwen3-Embedding-0.6B
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
Apr 20, 2026
Browser model files
onnx-community/Qwen3-Embedding-0.6B-ONNX
Files updated
Apr 3, 2026
Artifact revision
c25a394dd583836952667c12f008335071b3f43d

Input

TEXT → VECTORS

Sample result

Qwen3 Embedding 0.6B
Advanced settings 1
Scales each embedding vector to unit length. On makes vector lengths comparable for cosine similarity. Off preserves raw magnitudes only if supported; the listed fixed value must be retained.

Output

Explore what your text means

Generate vectors from one or more texts. Add multiple texts to compare their meaning.

Benchmark evidence

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

Qwen3 Embedding 0.6B

10 configurations tested
Passed10
Failed0
TestedSep 10, 2026
RuntimeTransformers.js 4.2.0

Quality criteria

Finite normalized 1024d vectors, distinct rows, and positive cosine exceeds negative by at least 0.1.

PrecisionDeviceResult
FP16GPU · WebGPUAcceleratedPassedFast
FP32GPU · WebGPUAcceleratedPassedFast
Q4F16GPU · WebGPUAcceleratedPassedModerate
Q4GPU · WebGPUAcceleratedPassedModerate
Q8GPU · WebGPUAcceleratedPassedModerate
FP32CPU · WASMPassedModerate
Q8CPU · WASMPassedModerate
FP16CPU · WASMPassedModerate
Q4F16CPU · WASMPassedSlow
Q4CPU · WASMPassedSlow
Test setup & sample details
Quality gate
Finite normalized 1024d vectors, distinct rows, and positive cosine exceeds negative by at least 0.1.
Tested
2026-09-10T10:07:48.000Z
Environment
Not recorded
Runtime
Transformers.js 4.2.0
Artifact revision
c25a394dd583836952667c12f008335071b3f43d
Fixture
{
  "file": "qwen3-embedding-0.6b.json",
  "url": "https://f005.backblazeb2.com/file/hubters-webai-public/assets/benchmarks/qwen3-embedding-0.6b.json",
  "size": 649,
  "sha1": "1de6f9cb512dd37a77b155454b595322c5eeff9f"
}
Rerun when
artifact revision change; runtime version change; fixture or quality gate change; browser/backend change
Per-case quality & failure details
[
  {
    "precision": "q8",
    "device": "wasm",
    "status": "pass",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": 66670.5,
      "inferenceMs": 1592.1999999284744,
      "totalMs": 69483.70000004768,
      "embeddingsPerSecond": 1.884,
      "grade": "very-slow"
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": 65435.89999997616,
      "inferenceMs": 1977,
      "totalMs": 68633.5,
      "embeddingsPerSecond": 1.517,
      "grade": "very-slow"
    }
  },
  {
    "precision": "q4",
    "device": "wasm",
    "status": "pass",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": 99191.29999995232,
      "inferenceMs": 2046.9000000953674,
      "totalMs": 102534.90000009537,
      "embeddingsPerSecond": 1.466,
      "grade": "very-slow"
    }
  },
  {
    "precision": "q4",
    "device": "webgpu",
    "status": "pass",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": 99817.80000007153,
      "inferenceMs": 700.6000000238419,
      "totalMs": 101760,
      "embeddingsPerSecond": 4.282,
      "grade": "slow"
    }
  },
  {
    "precision": "q4f16",
    "device": "wasm",
    "status": "pass",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": 63508.90000009537,
      "inferenceMs": 2042.7999999523163,
      "totalMs": 66920.60000002384,
      "embeddingsPerSecond": 1.469,
      "grade": "very-slow"
    }
  },
  {
    "precision": "q4f16",
    "device": "webgpu",
    "status": "pass",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": 67355.19999992847,
      "inferenceMs": 583,
      "totalMs": 69141.40000009537,
      "embeddingsPerSecond": 5.146,
      "grade": "slow"
    }
  },
  {
    "precision": "fp16",
    "device": "wasm",
    "status": "pass",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": 130781.79999995232,
      "inferenceMs": 1616.7999999523163,
      "totalMs": 133604.90000009537,
      "embeddingsPerSecond": 1.856,
      "grade": "very-slow"
    }
  },
  {
    "precision": "fp16",
    "device": "webgpu",
    "status": "pass",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": 134953.19999992847,
      "inferenceMs": 295.3000000715256,
      "totalMs": 136436.19999992847,
      "embeddingsPerSecond": 10.159,
      "grade": "good"
    }
  },
  {
    "precision": "fp32",
    "device": "wasm",
    "status": "pass",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": 245924.30000007153,
      "inferenceMs": 1547.6999999284744,
      "totalMs": 248731.89999997616,
      "embeddingsPerSecond": 1.938,
      "grade": "very-slow"
    }
  },
  {
    "precision": "fp32",
    "device": "webgpu",
    "status": "pass",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": 247768.39999997616,
      "inferenceMs": 377.2000000476837,
      "totalMs": 249519.10000002384,
      "embeddingsPerSecond": 7.953,
      "grade": "slow"
    }
  }
]

Qwen3 Embedding 4B

12 configurations tested
Passed12
Failed0
TestedNov 5, 2026
RuntimeTransformers.js 4.2.0

Quality criteria

{ "rows": 3, "dimensions": 2560, "positiveIndex": 1, "negativeIndex": 2, "minimumMargin": 0.1, "normTolerance": 0.02 }

PrecisionDeviceResult
Q8GPU · WebGPUAcceleratedPassedSlow
Q8GPU · WebGPUAcceleratedPassedSlow
Q8GPU · WebGPUAcceleratedPassedSlow
Q8GPU · WebGPUAcceleratedPassedSlow
Q8GPU · WebGPUAcceleratedPassedSlow
Q8GPU · WebGPUAcceleratedPassedSlow
Q8CPU · WASMPassedModerate
Q8CPU · WASMPassedModerate
Q8CPU · WASMPassedModerate
Q8CPU · WASMPassedModerate
Q8CPU · WASMPassedModerate
Q8CPU · WASMPassedSlow
Test setup & sample details
Quality gate
{ "rows": 3, "dimensions": 2560, "positiveIndex": 1, "negativeIndex": 2, "minimumMargin": 0.1, "normTolerance": 0.02 }
Tested
2026-11-05T12:47:18.162Z
Environment
Google Chrome 157.0.8103.39; macOS 27.2; Apple M3 Max; fresh isolated module workers; eight WASM threads; cross-origin isolated
Runtime
Transformers.js 4.2.0
Artifact revision
c214daf575fc82d981f9e60fb6563eb70f7cac8b
Fixture
{
  "file": "model-samples/qwen3-embedding-4b.json",
  "url": "https://assets.localai.tools/assets/benchmarks/5d2d3903303ae412cd292d906da63941b9672c7c7c3e074a50b1ede2f7b3dc58/qwen3-embedding-4b.json",
  "size": 659,
  "sha1": "92eb40b387097b2d3cc3fa3ce377f0c091735cd2"
}
Rerun when
artifact revision change; runtime version change; fixture or quality gate change; browser/backend change
Per-case quality & failure details
[
  {
    "precision": "q8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "finite": true,
      "shape": true,
      "normalized": true,
      "margin": 0.3855344169832811,
      "minimumMargin": 0.1,
      "passed": true
    },
    "runtime": {
      "loadMs": 19137.07499998808,
      "inferenceMs": 1911.4850000143051,
      "totalMs": 21049.269999980927,
      "embeddingsPerSecond": 1.5694603933473446,
      "grade": "very-slow",
      "wasmThreads": 8,
      "device": "wasm"
    }
  },
  {
    "precision": "q8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "finite": true,
      "shape": true,
      "normalized": true,
      "margin": 0.3855344169832811,
      "minimumMargin": 0.1,
      "passed": true
    },
    "runtime": {
      "loadMs": 19137.07499998808,
      "inferenceMs": 1772.3600000143051,
      "totalMs": 22821.805000007153,
      "embeddingsPerSecond": 1.6926583763884235,
      "grade": "very-slow",
      "wasmThreads": 8,
      "device": "wasm"
    }
  },
  {
    "precision": "q8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "finite": true,
      "shape": true,
      "normalized": true,
      "margin": 0.3855344169832811,
      "minimumMargin": 0.1,
      "passed": true
    },
    "runtime": {
      "loadMs": 19137.07499998808,
      "inferenceMs": 1857.824999988079,
      "totalMs": 24679.819999992847,
      "embeddingsPerSecond": 1.6147914900592089,
      "grade": "very-slow",
      "wasmThreads": 8,
      "device": "wasm"
    }
  },
  {
    "precision": "q8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "finite": true,
      "shape": true,
      "normalized": true,
      "margin": 0.3855344169832811,
      "minimumMargin": 0.1,
      "passed": true
    },
    "runtime": {
      "loadMs": 19137.07499998808,
      "inferenceMs": 2001.449999988079,
      "totalMs": 26681.67000001669,
      "embeddingsPerSecond": 1.4989132878752247,
      "grade": "very-slow",
      "wasmThreads": 8,
      "device": "wasm"
    }
  },
  {
    "precision": "q8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "finite": true,
      "shape": true,
      "normalized": true,
      "margin": 0.3855344169832811,
      "minimumMargin": 0.1,
      "passed": true
    },
    "runtime": {
      "loadMs": 19137.07499998808,
      "inferenceMs": 1813.2749999761581,
      "totalMs": 28495.12000000477,
      "embeddingsPerSecond": 1.6544649874064583,
      "grade": "very-slow",
      "wasmThreads": 8,
      "device": "wasm"
    }
  },
  {
    "precision": "q8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "finite": true,
      "shape": true,
      "normalized": true,
      "margin": 0.3855344169832811,
      "minimumMargin": 0.1,
      "passed": true
    },
    "runtime": {
      "loadMs": 19137.07499998808,
      "inferenceMs": 1874.0349999666214,
      "totalMs": 30369.310000002384,
      "embeddingsPerSecond": 1.6008238907242571,
      "grade": "very-slow",
      "wasmThreads": 8,
      "device": "wasm"
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "finite": true,
      "shape": true,
      "normalized": true,
      "margin": 0.39051471145899425,
      "minimumMargin": 0.1,
      "passed": true
    },
    "runtime": {
      "loadMs": 19995.055000007153,
      "inferenceMs": 2401.414999961853,
      "totalMs": 22397.139999985695,
      "embeddingsPerSecond": 1.2492634551077826,
      "grade": "very-slow",
      "wasmThreads": 8,
      "device": "webgpu"
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "finite": true,
      "shape": true,
      "normalized": true,
      "margin": 0.39051471145899425,
      "minimumMargin": 0.1,
      "passed": true
    },
    "runtime": {
      "loadMs": 19995.055000007153,
      "inferenceMs": 2384.8400000333786,
      "totalMs": 24782.15499997139,
      "embeddingsPerSecond": 1.2579460257115829,
      "grade": "very-slow",
      "wasmThreads": 8,
      "device": "webgpu"
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "finite": true,
      "shape": true,
      "normalized": true,
      "margin": 0.39051471145899425,
      "minimumMargin": 0.1,
      "passed": true
    },
    "runtime": {
      "loadMs": 19995.055000007153,
      "inferenceMs": 2225.214999973774,
      "totalMs": 27007.569999992847,
      "embeddingsPerSecond": 1.3481843327657586,
      "grade": "very-slow",
      "wasmThreads": 8,
      "device": "webgpu"
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "finite": true,
      "shape": true,
      "normalized": true,
      "margin": 0.39051471145899425,
      "minimumMargin": 0.1,
      "passed": true
    },
    "runtime": {
      "loadMs": 19995.055000007153,
      "inferenceMs": 2518.699999988079,
      "totalMs": 29526.634999990463,
      "embeddingsPerSecond": 1.1910906420034935,
      "grade": "very-slow",
      "wasmThreads": 8,
      "device": "webgpu"
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "finite": true,
      "shape": true,
      "normalized": true,
      "margin": 0.39051471145899425,
      "minimumMargin": 0.1,
      "passed": true
    },
    "runtime": {
      "loadMs": 19995.055000007153,
      "inferenceMs": 2767.574999988079,
      "totalMs": 32294.37000000477,
      "embeddingsPerSecond": 1.083981463921636,
      "grade": "very-slow",
      "wasmThreads": 8,
      "device": "webgpu"
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "finite": true,
      "shape": true,
      "normalized": true,
      "margin": 0.39051471145899425,
      "minimumMargin": 0.1,
      "passed": true
    },
    "runtime": {
      "loadMs": 19995.055000007153,
      "inferenceMs": 2523.1800000071526,
      "totalMs": 34817.99500000477,
      "embeddingsPerSecond": 1.1889758162285273,
      "grade": "very-slow",
      "wasmThreads": 8,
      "device": "webgpu"
    }
  }
]

Something not quite right?

Tell us what happened with Qwen3 Embedding. 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.