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.

Florence 2

Read text, describe images, and explore regions locally.

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

Describe an image, read visible text, or locate objects and phrases using the selected vision task. Returns text and task-specific regions or labels; available output depends on the task you choose.

Created by
Microsoft
License
MIT
Model released
Unknown
Source repository created
Jun 15, 2024
Model weights · Q4
333.2 MB

Best for

  • Generate short or detailed descriptions of photographs.
  • Extract visible text or spatially grounded labels from images.
  • Prototype object detection, region proposal, and phrase-grounding interfaces.

Keep in mind

  • Outputs can be incomplete, incorrect, biased, or hallucinated and require review.
  • Small, blurry, rotated, stylized, low-contrast, or dense text can reduce OCR accuracy.
  • Bounding boxes and region labels can miss, merge, or mislocalize objects.
  • The model is English-centric and must not be the sole basis for safety-critical, accessibility, or identity decisions.

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
microsoft/Florence-2-base-ft
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
Aug 4, 2025
Browser model files
onnx-community/Florence-2-base-ft
Files updated
May 8, 2025
Artifact revision
e88a44eaf3791a35eae0c5a47b3dbcd36e67eb6f

Input

PNG / JPG / WEBP

Sample result

Florence-2 Base FT — Image + Text

Settings

Florence task prompt: caption, detailed caption, more-detailed caption, object detection, dense-region caption, region proposal, OCR, OCR with regions, or caption-to-phrase grounding. Choose a task matching the desired output. Caption tasks produce descriptions, OCR reads text, detection/grounding locates regions when offered. For speech, transcribe preserves the spoken language and translate outputs English; unsupported task values are not available.
Limits the number of newly generated output tokens. Lower values finish sooner but can cut answers short. Higher values allow longer answers and can use more time and memory; they do not guarantee better answers.
Advanced settings 1
Adds an image with task text or spatial regions while retaining the complete native result. On adds a downloadable visual preview. Off returns only native task fields and avoids presentation work.

Output

Your document starts here

Choose a page image or try the sample to see extracted text and structure.

Benchmark evidence

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

Florence-2 Base FT — Image + Text

17 configurations tested
Passed13
Failed4
TestedSep 12, 2026
RuntimeTransformers.js 4.2.0

Quality criteria

Non-empty, non-repetitive caption that identifies a person in the checked-in portrait; streamed text must equal non-streamed text.

PrecisionDeviceResult
Q4GPU · WebGPUAcceleratedPassedModerate
FP32GPU · WebGPUAcceleratedPassedModerate
FP32GPU · WebGPUAcceleratedPassedModerate
BNB4GPU · WebGPUAcceleratedPassedVery slow
UINT8GPU · WebGPUAcceleratedPassedVery slow
Q8GPU · WebGPUAcceleratedPassedVery slow
INT8GPU · WebGPUAcceleratedPassedVery slow
Q8CPU · WASMPassedVery slow
UINT8CPU · WASMPassedVery slow
BNB4CPU · WASMPassedVery slow
FP32CPU · WASMPassedVery slow
INT8CPU · WASMPassedVery slow
Q4CPU · WASMPassedVery slow
FP16CPU · WASMFailed—
FP16GPU · WebGPUFailed—
Q4F16CPU · WASMFailed—
Q4F16GPU · WebGPUFailed—
Test setup & sample details
Quality gate
Non-empty, non-repetitive caption that identifies a person in the checked-in portrait; streamed text must equal non-streamed text.
Tested
2026-09-12T18:20:09.043Z
Environment
Isolated headless Chrome on macOS; Transformers.js 4.2.0
Runtime
Transformers.js 4.2.0
Artifact revision
e88a44eaf3791a35eae0c5a47b3dbcd36e67eb6f
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",
  "licenseUrl": "https://www.pexels.com/license/"
}
Rerun when
artifact revision change; runtime version change; fixture or quality-gate change; browser/backend behavior change
Per-case quality & failure details
[
  {
    "precision": "fp32",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "identifiesPerson": true,
      "finite": true,
      "nonRepetitive": true,
      "streamMatches": true,
      "wordCount": 12,
      "grade": "caption-pass"
    },
    "runtime": {
      "loadMs": 108049.29999999702,
      "inferenceMs": 11079.20000000298,
      "totalMs": 131064.69999999553,
      "grade": "very-slow",
      "streamInferenceMs": 10583.89999999851,
      "imagesPerSecond": 0.09025922449271888
    }
  },
  {
    "precision": "fp32",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "identifiesPerson": true,
      "finite": true,
      "nonRepetitive": true,
      "streamMatches": true,
      "wordCount": 12,
      "grade": "caption-pass"
    },
    "runtime": {
      "loadMs": 108948.60000000149,
      "inferenceMs": 617.6999999955297,
      "totalMs": 111278,
      "grade": "excellent",
      "streamInferenceMs": 454.20000000298023,
      "imagesPerSecond": 1.6189088554431554
    }
  },
  {
    "precision": "fp16",
    "device": "wasm",
    "status": "fail",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": null,
      "inferenceMs": null,
      "totalMs": 32635.5,
      "grade": "failed"
    },
    "failure": "Error: Can't create a session. ERROR_CODE: 1, ERROR_MESSAGE: /mnt/vss/_work/1/s/onnxruntime/core/graph/graph_utils.cc:30 int onnxruntime::graph_utils::GetIndexFromName(const Node &, const std::string &, bool) itr != node_args.end() was false. Attempting to get index by a name which does not exist:InsertedPrecisionFreeCast_/layers.5/self_attn_layer_norm/Constant_output_0for node: /layernorm_embedding/Mul/SimplifiedLayerNormFusion/\n"
  },
  {
    "precision": "fp16",
    "device": "webgpu",
    "status": "fail",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": null,
      "inferenceMs": null,
      "totalMs": 60980,
      "grade": "failed"
    },
    "failure": "Error: Can't create a session. ERROR_CODE: 1, ERROR_MESSAGE: /mnt/vss/_work/1/s/onnxruntime/core/graph/graph.cc:1491 void onnxruntime::Graph::InitializeStateFromModelFileGraphProto() This is an invalid model. Subgraph output (logits) is an outer scope value being returned directly. Please update the model to add an Identity node between the outer scope value and the subgraph output.\n"
  },
  {
    "precision": "int8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "identifiesPerson": true,
      "finite": true,
      "nonRepetitive": true,
      "streamMatches": true,
      "wordCount": 12,
      "grade": "caption-pass"
    },
    "runtime": {
      "loadMs": 33050.10000000149,
      "inferenceMs": 11720.39999999851,
      "totalMs": 57501.60000000149,
      "grade": "very-slow",
      "streamInferenceMs": 11499.20000000298,
      "imagesPerSecond": 0.0853213200914753
    }
  },
  {
    "precision": "int8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "identifiesPerson": true,
      "finite": true,
      "nonRepetitive": true,
      "streamMatches": true,
      "wordCount": 12,
      "grade": "caption-pass"
    },
    "runtime": {
      "loadMs": 32247.70000000298,
      "inferenceMs": 11916.89999999851,
      "totalMs": 57049.30000000447,
      "grade": "very-slow",
      "streamInferenceMs": 11538.80000000447,
      "imagesPerSecond": 0.08391444083613397
    }
  },
  {
    "precision": "uint8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "identifiesPerson": true,
      "finite": true,
      "nonRepetitive": true,
      "streamMatches": true,
      "wordCount": 12,
      "grade": "caption-pass"
    },
    "runtime": {
      "loadMs": 31642.10000000149,
      "inferenceMs": 10279,
      "totalMs": 53303.39999999851,
      "grade": "very-slow",
      "streamInferenceMs": 10117.20000000298,
      "imagesPerSecond": 0.09728572818367545
    }
  },
  {
    "precision": "uint8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "identifiesPerson": true,
      "finite": true,
      "nonRepetitive": true,
      "streamMatches": true,
      "wordCount": 12,
      "grade": "caption-pass"
    },
    "runtime": {
      "loadMs": 31365.29999999702,
      "inferenceMs": 10366.5,
      "totalMs": 53327.5,
      "grade": "very-slow",
      "streamInferenceMs": 10168.30000000447,
      "imagesPerSecond": 0.0964645733854242
    }
  },
  {
    "precision": "q8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "identifiesPerson": true,
      "finite": true,
      "nonRepetitive": true,
      "streamMatches": true,
      "wordCount": 12,
      "grade": "caption-pass"
    },
    "runtime": {
      "loadMs": 30896.20000000298,
      "inferenceMs": 10247.89999999851,
      "totalMs": 52584.10000000149,
      "grade": "very-slow",
      "streamInferenceMs": 10127.60000000149,
      "imagesPerSecond": 0.0975809678080529
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "identifiesPerson": true,
      "finite": true,
      "nonRepetitive": true,
      "streamMatches": true,
      "wordCount": 12,
      "grade": "caption-pass"
    },
    "runtime": {
      "loadMs": 32917.60000000149,
      "inferenceMs": 10382.60000000149,
      "totalMs": 54766.90000000596,
      "grade": "very-slow",
      "streamInferenceMs": 10149.89999999851,
      "imagesPerSecond": 0.09631498853850254
    }
  },
  {
    "precision": "q4",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "identifiesPerson": true,
      "finite": true,
      "nonRepetitive": true,
      "streamMatches": true,
      "wordCount": 12,
      "grade": "caption-pass"
    },
    "runtime": {
      "loadMs": 42147.69999999553,
      "inferenceMs": 11727.30000000447,
      "totalMs": 66847.5,
      "grade": "very-slow",
      "streamInferenceMs": 11659.39999999851,
      "imagesPerSecond": 0.08527111952449573
    }
  },
  {
    "precision": "q4",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "identifiesPerson": true,
      "finite": true,
      "nonRepetitive": true,
      "streamMatches": true,
      "wordCount": 12,
      "grade": "caption-pass"
    },
    "runtime": {
      "loadMs": 41324.70000000298,
      "inferenceMs": 543.9000000059605,
      "totalMs": 43678.60000000149,
      "grade": "excellent",
      "streamInferenceMs": 440,
      "imagesPerSecond": 1.8385732671245472
    }
  },
  {
    "precision": "q4f16",
    "device": "wasm",
    "status": "fail",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": null,
      "inferenceMs": null,
      "totalMs": 19000.10000000149,
      "grade": "failed"
    },
    "failure": "Error: Can't create a session. ERROR_CODE: 1, ERROR_MESSAGE: /mnt/vss/_work/1/s/onnxruntime/core/graph/graph_utils.cc:30 int onnxruntime::graph_utils::GetIndexFromName(const Node &, const std::string &, bool) itr != node_args.end() was false. Attempting to get index by a name which does not exist:InsertedPrecisionFreeCast_/layers.5/self_attn_layer_norm/Constant_output_0for node: /layernorm_embedding/Mul/SimplifiedLayerNormFusion/\n"
  },
  {
    "precision": "q4f16",
    "device": "webgpu",
    "status": "fail",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": null,
      "inferenceMs": null,
      "totalMs": 47430.10000000149,
      "grade": "failed"
    },
    "failure": "Error: Can't create a session. ERROR_CODE: 1, ERROR_MESSAGE: /mnt/vss/_work/1/s/onnxruntime/core/graph/graph.cc:1491 void onnxruntime::Graph::InitializeStateFromModelFileGraphProto() This is an invalid model. Subgraph output (logits) is an outer scope value being returned directly. Please update the model to add an Identity node between the outer scope value and the subgraph output.\n"
  },
  {
    "precision": "bnb4",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "identifiesPerson": true,
      "finite": true,
      "nonRepetitive": true,
      "streamMatches": true,
      "wordCount": 12,
      "grade": "caption-pass"
    },
    "runtime": {
      "loadMs": 40501.5,
      "inferenceMs": 10628.29999999702,
      "totalMs": 63013.19999999553,
      "grade": "very-slow",
      "streamInferenceMs": 10512.30000000447,
      "imagesPerSecond": 0.09408842430118461
    }
  },
  {
    "precision": "bnb4",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "identifiesPerson": true,
      "finite": true,
      "nonRepetitive": true,
      "streamMatches": true,
      "wordCount": 12,
      "grade": "caption-pass"
    },
    "runtime": {
      "loadMs": 40026.30000000447,
      "inferenceMs": 10309.29999999702,
      "totalMs": 61939.60000000149,
      "grade": "very-slow",
      "streamInferenceMs": 10240.60000000149,
      "imagesPerSecond": 0.09699979630045581
    }
  },
  {
    "precision": "fp32",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "method": "Normalized exact authored printed text"
    },
    "runtime": {
      "loadMs": 24522.519999980927,
      "inferenceMs": 554.3650000095367,
      "totalMs": 25079.469999969006,
      "pagesPerSecond": 1.8038656841301255,
      "grade": "fast"
    }
  }
]

Florence-2 Base — Image + Text

5 configurations tested
Passed5
Failed0
TestedNov 5, 2026
RuntimeTransformers.js 4.2.0

Quality criteria

Portrait caption identifies a person, at least 5 words, no repeated four-word phrase; streaming matches. Speed gate: inference <=30sec. Narrow validity only, not an app performance benchmark.

PrecisionDeviceResult
Q4GPU · WebGPUAcceleratedPassedModerate
Q8GPU · WebGPUAcceleratedPassedSlow
Q8GPU · WebGPUAcceleratedPassedVery slow
Q8CPU · WASMPassedSlow
Q4CPU · WASMPassedVery slow
Test setup & sample details
Quality gate
Portrait caption identifies a person, at least 5 words, no repeated four-word phrase; streaming matches. Speed gate: inference <=30sec. Narrow validity only, not an app performance benchmark.
Tested
2026-11-05T03:17:55.904Z
Environment
Isolated headless Chrome on macOS; Transformers.js 4.2.0
Runtime
Transformers.js 4.2.0
Artifact revision
d59e079711c57174f29265539fb4cc9f0f335916
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",
  "licenseUrl": "https://www.pexels.com/license/"
}
Rerun when
artifact revision change; runtime version change; fixture or quality-gate change; browser/backend behavior change
Per-case quality & failure details
[
  {
    "precision": "q8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "subject": true,
      "nonempty": true,
      "nonrepetitive": true,
      "streamMatches": true,
      "incremental": true,
      "validityScore": 1,
      "grade": "caption-validity-pass"
    },
    "runtime": {
      "loadMs": 29914.71500003338,
      "inferenceMs": 3697.939999997616,
      "totalMs": 38758.82499998808,
      "grade": "good",
      "streamMs": 3812.7199999690056,
      "imagesPerSecond": 0.2704208288940991
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "subject": true,
      "nonempty": true,
      "nonrepetitive": true,
      "streamMatches": true,
      "incremental": true,
      "validityScore": 1,
      "grade": "caption-validity-pass"
    },
    "runtime": {
      "loadMs": 30090.280000030994,
      "inferenceMs": 7251.6449999809265,
      "totalMs": 46350.330000042915,
      "grade": "good",
      "streamMs": 7437.045000016689,
      "imagesPerSecond": 0.13789974550638237
    }
  },
  {
    "precision": "q4",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "subject": true,
      "nonempty": true,
      "nonrepetitive": true,
      "streamMatches": true,
      "incremental": true,
      "validityScore": 1,
      "grade": "caption-validity-pass"
    },
    "runtime": {
      "loadMs": 84935.07500004768,
      "inferenceMs": 6967.72000002861,
      "totalMs": 100406.46000003815,
      "grade": "good",
      "streamMs": 7119.724999964237,
      "imagesPerSecond": 0.1435189703369099
    }
  },
  {
    "precision": "q4",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "subject": true,
      "nonempty": true,
      "nonrepetitive": true,
      "streamMatches": true,
      "incremental": true,
      "validityScore": 1,
      "grade": "caption-validity-pass"
    },
    "runtime": {
      "loadMs": 36082.18000000715,
      "inferenceMs": 1354.539999961853,
      "totalMs": 40027.12999999523,
      "grade": "excellent",
      "streamMs": 1061.4350000023842,
      "imagesPerSecond": 0.7382580064288705
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "method": "Normalized exact authored printed text"
    },
    "runtime": {
      "loadMs": 7700.259999990463,
      "inferenceMs": 3977.75,
      "totalMs": 11680.139999985695,
      "pagesPerSecond": 0.251398403620137,
      "grade": "good"
    }
  }
]

Florence-2 Large — Image + Text

5 configurations tested
Passed5
Failed0
TestedNov 5, 2026
RuntimeTransformers.js 4.2.0

Quality criteria

Portrait caption identifies a person, at least 5 words, no repeated four-word phrase; streaming matches. Speed gate: inference <=30sec. Narrow validity only, not an app performance benchmark.

PrecisionDeviceResult
Q4GPU · WebGPUAcceleratedPassedSlow
Q8GPU · WebGPUAcceleratedPassedVery slow
Q8GPU · WebGPUAcceleratedPassedVery slow
Q8CPU · WASMPassedVery slow
Q4CPU · WASMPassedVery slow
Test setup & sample details
Quality gate
Portrait caption identifies a person, at least 5 words, no repeated four-word phrase; streaming matches. Speed gate: inference <=30sec. Narrow validity only, not an app performance benchmark.
Tested
2026-11-05T03:26:42.277Z
Environment
Isolated headless Chrome on macOS; Transformers.js 4.2.0
Runtime
Transformers.js 4.2.0
Artifact revision
b7751295927207b2a8c2f354af8baadea5b0404b
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",
  "licenseUrl": "https://www.pexels.com/license/"
}
Rerun when
artifact revision change; runtime version change; fixture or quality-gate change; browser/backend behavior change
Per-case quality & failure details
[
  {
    "precision": "q8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "subject": true,
      "nonempty": true,
      "nonrepetitive": true,
      "streamMatches": true,
      "incremental": true,
      "validityScore": 1,
      "grade": "caption-validity-pass"
    },
    "runtime": {
      "loadMs": 142940.125,
      "inferenceMs": 12648.930000007153,
      "totalMs": 169623.79000002146,
      "grade": "slow",
      "streamMs": 12523.370000004768,
      "imagesPerSecond": 0.07905807052449769
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "subject": true,
      "nonempty": true,
      "nonrepetitive": true,
      "streamMatches": true,
      "incremental": true,
      "validityScore": 1,
      "grade": "caption-validity-pass"
    },
    "runtime": {
      "loadMs": 83950.55500000715,
      "inferenceMs": 19954.589999973774,
      "totalMs": 125761.875,
      "grade": "slow",
      "streamMs": 20331.855000019073,
      "imagesPerSecond": 0.05011378334515088
    }
  },
  {
    "precision": "q4",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "subject": true,
      "nonempty": true,
      "nonrepetitive": true,
      "streamMatches": true,
      "incremental": true,
      "validityScore": 1,
      "grade": "caption-validity-pass"
    },
    "runtime": {
      "loadMs": 104481.74500000477,
      "inferenceMs": 18455.185000002384,
      "totalMs": 142346.3299999833,
      "grade": "slow",
      "streamMs": 17936.235000014305,
      "imagesPerSecond": 0.05418531431680965
    }
  },
  {
    "precision": "q4",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "subject": true,
      "nonempty": true,
      "nonrepetitive": true,
      "streamMatches": true,
      "incremental": true,
      "validityScore": 1,
      "grade": "caption-validity-pass"
    },
    "runtime": {
      "loadMs": 80853.43999999762,
      "inferenceMs": 2049.089999973774,
      "totalMs": 86242.36500000954,
      "grade": "good",
      "streamMs": 1902.7250000238419,
      "imagesPerSecond": 0.48802151199449456
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "method": "Normalized exact authored printed text"
    },
    "runtime": {
      "loadMs": 23044.844999969006,
      "inferenceMs": 11645.055000007153,
      "totalMs": 34691.90999996662,
      "pagesPerSecond": 0.0858733599797842,
      "grade": "slow"
    }
  }
]

Florence-2 Large FT — Image + Text

5 configurations tested
Passed5
Failed0
TestedNov 5, 2026
RuntimeTransformers.js 4.2.0

Quality criteria

Portrait caption identifies a person, at least 5 words, no repeated four-word phrase; streaming matches. Speed gate: inference <=30sec. Narrow validity only, not an app performance benchmark.

PrecisionDeviceResult
Q4GPU · WebGPUAcceleratedPassedModerate
Q8GPU · WebGPUAcceleratedPassedVery slow
Q8GPU · WebGPUAcceleratedPassedVery slow
Q8CPU · WASMPassedVery slow
Q4CPU · WASMPassedVery slow
Test setup & sample details
Quality gate
Portrait caption identifies a person, at least 5 words, no repeated four-word phrase; streaming matches. Speed gate: inference <=30sec. Narrow validity only, not an app performance benchmark.
Tested
2026-11-05T03:33:38.627Z
Environment
Isolated headless Chrome on macOS; Transformers.js 4.2.0
Runtime
Transformers.js 4.2.0
Artifact revision
04ace74913c28d7ec94af32cee5c111a10126b6f
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",
  "licenseUrl": "https://www.pexels.com/license/"
}
Rerun when
artifact revision change; runtime version change; fixture or quality-gate change; browser/backend behavior change
Per-case quality & failure details
[
  {
    "precision": "q8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "subject": true,
      "nonempty": true,
      "nonrepetitive": true,
      "streamMatches": true,
      "incremental": true,
      "validityScore": 1,
      "grade": "caption-validity-pass"
    },
    "runtime": {
      "loadMs": 81724.18999999762,
      "inferenceMs": 12233.209999978542,
      "totalMs": 107008.24000000954,
      "grade": "slow",
      "streamMs": 11597.590000033379,
      "imagesPerSecond": 0.08174469333901356
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "subject": true,
      "nonempty": true,
      "nonrepetitive": true,
      "streamMatches": true,
      "incremental": true,
      "validityScore": 1,
      "grade": "caption-validity-pass"
    },
    "runtime": {
      "loadMs": 81957.78999996185,
      "inferenceMs": 16025.120000004768,
      "totalMs": 116600.02500003576,
      "grade": "slow",
      "streamMs": 17092.929999947548,
      "imagesPerSecond": 0.06240202881474226
    }
  },
  {
    "precision": "q4",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "subject": true,
      "nonempty": true,
      "nonrepetitive": true,
      "streamMatches": true,
      "incremental": true,
      "validityScore": 1,
      "grade": "caption-validity-pass"
    },
    "runtime": {
      "loadMs": 75964.21000003815,
      "inferenceMs": 14791,
      "totalMs": 107017.77999997139,
      "grade": "slow",
      "streamMs": 14793.534999966621,
      "imagesPerSecond": 0.06760868095463457
    }
  },
  {
    "precision": "q4",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "subject": true,
      "nonempty": true,
      "nonrepetitive": true,
      "streamMatches": true,
      "incremental": true,
      "validityScore": 1,
      "grade": "caption-validity-pass"
    },
    "runtime": {
      "loadMs": 78526.94499999285,
      "inferenceMs": 1568.6899999976158,
      "totalMs": 83203.67500001192,
      "grade": "excellent",
      "streamMs": 1584.1399999856949,
      "imagesPerSecond": 0.6374745807020634
    }
  },
  {
    "precision": "q8",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "method": "Normalized exact authored printed text"
    },
    "runtime": {
      "loadMs": 23819.69999998808,
      "inferenceMs": 11467.680000007153,
      "totalMs": 35289.88499999046,
      "pagesPerSecond": 0.08720159613796132,
      "grade": "slow"
    }
  }
]

Something not quite right?

Tell us what happened with Florence 2. 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.