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.

PaddleOCR-VL 1.6

Read text from PNG/JPEG document regions or locally rasterized PDF pages with native PaddleOCR-VL 1.6. Returns complete raw OCR text and per-page results.

Desktop inferencePrivate · Runs on your computer
About this model

Read text from PNG/JPEG document regions or locally rasterized PDF pages with native PaddleOCR-VL 1.6. Returns complete raw OCR text and per-page results.

License
Apache-2.0
Model released
May 28, 2026
Source repository created
May 27, 2026
Model weights · FP32
1.9 GB

Best for

  • Local document-region text recognition with manual review
  • Reading simple scanned pages and raster PDFs
  • Reproducible OCR experiments on authored documents

Keep in mind

  • OCR can omit, duplicate, misread or hallucinate text; verify names, numbers and punctuation against originals. Not legal/medical/financial/identity or redaction ground truth.
  • This adapter uses the official Transformers element-recognition OCR prompt. It does not implement the complete PaddleOCR layout-detection/refinement/page-parsing pipeline or guarantee reading order.
  • Table content may be read as text; no structured table, formula, chart, seal or spotting output is certified. Those task prompts require separate quality gates.
  • Upstream multilingual capability is not local quality certification: current suite covers authored English documents only. Handwriting, tiny text, poor scans, complex columns and other languages remain unverified.
  • PDF pages are rasterized with PDFium and processed in order, without embedded-text extraction. Invalid/truncated/encrypted files fail validation. Memory/time increase with pages, resolution and output tokens.
  • Per-page output budget may truncate text. Inspect outputLimitReached and allPagesReachedEos; partial text is not complete extraction.
  • Only measured FP32 CPU/Metal pairs are selectable. FP16/BF16 are isolated probes only. Other Mac hardware/OS and Windows providers remain unverified.
  • Model weights and private pinned runtime download on request. Current normal provisioning requires Apple Silicon and macOS14+.
  • One official 1.6 0.9B checkpoint; GGUF is a weight/runtime format and earlier1.5/original are prior versions, not smaller1.6 models.

Processed on your deviceYour input is processed locally in the desktop app. Model and runtime files download from external hosts.

Target platforms · verification required

  • macos · aarch64, x86_64
  • windows · x86_64

Recommended configuration

FP32 · mps (verified platform only)

Sources & technical details
Original model
https://huggingface.co/PaddlePaddle/PaddleOCR-VL-1.6
Release verification
Verified
Release evidence
Exact original 1.6 checkpoint release stated in official model-card news dated2026.05.28; distinct from repository creation, paper and subsequent update.
Source updated
Aug 8, 2026
Native model files
https://huggingface.co/PaddlePaddle/PaddleOCR-VL-1.6
Files updated
Aug 8, 2026
Artifact revision
c5630abae1d940eafe0697512a0325494b02ab42

Desktop app

Sample result

PaddleOCR-VL 1.6 · 0.9B — Image to Text

Run PaddleOCR-VL 1.6 on your computer

Download the desktop app to load this model, provide input and view results locally.

Public desktop downloads are coming soon.

Device requirements

macOS Apple SiliconCPU / FP32Tested16.0 GiBRAM estimate
macOS Apple SiliconMPS / FP32Tested16.0 GiBRAM estimate
macOS Apple SiliconMPS / FP16Unverified16.0 GiBRAM estimate
macOS Apple SiliconMPS / BF16Unverified16.0 GiBRAM estimate
Windows Intel / AMDCPU / FP32Unverified16.0 GiBRAM estimate
macOS Intel / AMDCPU / FP32Unverified16.0 GiBRAM estimate
If your device falls short

MemoryAdvisory only: close other apps, reduce image/tile/token/batch settings or choose a smaller verified model/precision. CPU can need more RAM than GPU.

GPUIf the selected GPU runtime fails, existing CPU fallback may recover when enabled, but CPU has its own memory/runtime needs.

PlatformUse a compatible OS/architecture/runtime or another model. These estimates do not add hardware detection or block runs.

Benchmark evidence

Saved desktop measurements apply to the recorded hardware and runtime. Fallback cases report the device that actually ran; they do not prove GPU support on the requested device.

PaddleOCR-VL 1.6 · 0.9B — Image to Text

35 configurations tested
Passed35
Failed0
TestedNov 2, 2026
Runtimetransformers 5.22.1

Quality criteria

No explicit quality gate recorded. Review per-case evidence; full certification is pending.

PrecisionDeviceResultPlatformInference / load
FP32mpsAcceleratedPassedFastmacos · aarch641736 ms / 0 ms
FP32mpsAcceleratedPassedFastmacos · aarch642235 ms / 0 ms
FP32mpsAcceleratedPassedFastmacos · aarch642305 ms / 0 ms
FP32mpsAcceleratedPassedFastmacos · aarch642313 ms / 0 ms
FP32mpsAcceleratedPassedFastmacos · aarch642322 ms / 0 ms
FP32mpsAcceleratedPassedFastmacos · aarch642337 ms / 0 ms
FP32mpsAcceleratedPassedFastmacos · aarch642349 ms / 0 ms
FP32mpsAcceleratedPassedFastmacos · aarch642553 ms / 711 ms
BF16mpsAcceleratedPassedFastmacos · aarch642654 ms / 619 ms
FP32mpsAcceleratedPassedFastmacos · aarch642660 ms / 654 ms
FP16mpsAcceleratedPassedFastmacos · aarch642788 ms / 568 ms
FP32mpsAcceleratedPassedFastmacos · aarch643066 ms / 704 ms
FP32mpsAcceleratedPassedFastmacos · aarch646620 ms / 0 ms
FP32cpuPassedFastmacos · aarch644305 ms / 0 ms
FP32cpuPassedFastmacos · aarch645407 ms / 0 ms
FP32cpuPassedFastmacos · aarch645484 ms / 359 ms
FP32cpuPassedFastmacos · aarch645535 ms / 357 ms
FP32cpuPassedFastmacos · aarch645563 ms / 369 ms
FP32cpuPassedFastmacos · aarch645567 ms / 0 ms
FP32cpuPassedFastmacos · aarch645573 ms / 0 ms
FP32cpuPassedFastmacos · aarch645573 ms / 0 ms
FP32cpuPassedFastmacos · aarch645583 ms / 0 ms
FP32cpuPassedFastmacos · aarch645624 ms / 0 ms
FP32cpuPassedFastmacos · aarch649835 ms / 0 ms
FP32cpuPassedFastmacos · aarch6411078 ms / 0 ms
FP32cpuPassedFastmacos · aarch6411238 ms / 0 ms
FP32cpuPassedFastmacos · aarch6411259 ms / 0 ms
FP32cpuPassedFastmacos · aarch6411263 ms / 0 ms
FP32cpuPassedFastmacos · aarch6411268 ms / 0 ms
FP32cpuPassedFastmacos · aarch6411271 ms / 0 ms
FP32cpuPassedFastmacos · aarch6411280 ms / 401 ms
FP32cpuPassedFastmacos · aarch6411318 ms / 388 ms
FP32cpuPassedFastmacos · aarch6411362 ms / 373 ms
FP32cpuPassedFastmacos · aarch6414197 ms / 0 ms
FP32cpuPassedFastmacos · aarch6431051 ms / 0 ms
Test setup & sample details
Quality gate
No explicit quality gate recorded. Review the per-case quality evidence; full certification is pending.
Tested
2026-11-02T19:09:06.390642+00:00
Environment
{ "hardware": "Apple M3 Max", "os": "macOS-27.2-arm64-arm-64bit-Mach-O", "python": "3.13.18 (app-owned private development venv)", "runtime": "Transformers5.22.1/Torch2.10.0/PDFium5.14.0" }
Runtime
transformers 5.22.1
Artifact revision
c5630abae1d940eafe0697512a0325494b02ab42
Fixture
{
  "file": "model-samples/desktop/docling/receipt.png",
  "size": 25773,
  "sha1": "05f77f558dde3e17f3541910abeecde08c193cd5",
  "sha256": "cc0123c5e9e48de091ce0d3462d267b2485295096206353c9bf4beeec594ca0e",
  "url": "https://assets.localai.tools/samples/v1/files/cc0123c5e9e48de091ce0d3462d267b2485295096206353c9bf4beeec594ca0e/receipt.png"
}
Rerun when
artifact revision/checksums; Transformers/Torch/PDFium/runtime/driver/backend; OCR prompt/preprocessing/PDF rendering/output config; fixtures/expected text/quality and usability gates
Per-case quality & failure details
[
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 400.645000001532,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 11324.802291986998,
      "inferenceMs": 11279.520832991693,
      "pagesPerSecond": 0.08865624832883683,
      "tokensPerSecond": 4.521468664770678,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 11725.44729198853,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 11091.24416600389,
      "inferenceMs": 11077.8712080064,
      "pagesPerSecond": 0.09027005109765691,
      "tokensPerSecond": 4.062152299394561,
      "outputTokens": 45,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 11091.24416600389,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 31083.505417002016,
      "inferenceMs": 31051.173749990994,
      "pagesPerSecond": 0.06440980351026442,
      "tokensPerSecond": 2.18993331934899,
      "outputTokens": 68,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "application/pdf",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 31083.505417002016,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 0,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 0,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 9847.740416997112,
      "inferenceMs": 9834.564666991355,
      "pagesPerSecond": 0.10168218257350944,
      "tokensPerSecond": 0.10168218257350944,
      "outputTokens": 1,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 9847.740416997112,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 11250.644375002594,
      "inferenceMs": 11237.856709005428,
      "pagesPerSecond": 0.08898493955690435,
      "tokensPerSecond": 4.538231917402122,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 11250.644375002594,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 11283.710542003973,
      "inferenceMs": 11270.655750005972,
      "pagesPerSecond": 0.08872598207069453,
      "tokensPerSecond": 4.525025085605421,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 11283.710542003973,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 11272.106457996415,
      "inferenceMs": 11258.873458005837,
      "pagesPerSecond": 0.08881883287256692,
      "tokensPerSecond": 4.529760476500913,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 11272.106457996415,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 11276.078791997861,
      "inferenceMs": 11263.19120900007,
      "pagesPerSecond": 0.08878478412059015,
      "tokensPerSecond": 4.528023990150098,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 11276.078791997861,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 11281.033000006573,
      "inferenceMs": 11267.897458004882,
      "pagesPerSecond": 0.0887477014879635,
      "tokensPerSecond": 4.526132775886139,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 11281.033000006573,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 387.8166670037899,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 11339.594042001409,
      "inferenceMs": 11317.977541999426,
      "pagesPerSecond": 0.08835500832981338,
      "tokensPerSecond": 4.506105424820483,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 11727.410709005198,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 373.2784169988008,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 11383.419749996392,
      "inferenceMs": 11361.55804200098,
      "pagesPerSecond": 0.08801609746684722,
      "tokensPerSecond": 4.488820970809208,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 11756.698166995193,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 357.00574998918455,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 5549.011916999007,
      "inferenceMs": 5535.161749998224,
      "pagesPerSecond": 0.1806631938082606,
      "tokensPerSecond": 9.21382288422129,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 5906.017666988191,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 5412.052917003166,
      "inferenceMs": 5406.638500004192,
      "pagesPerSecond": 0.18495780696253775,
      "tokensPerSecond": 8.323101313314199,
      "outputTokens": 45,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 5412.052917003166,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 14211.072959005833,
      "inferenceMs": 14197.4435000011,
      "pagesPerSecond": 0.14087043206052166,
      "tokensPerSecond": 4.789594690057737,
      "outputTokens": 68,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "application/pdf",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 14211.072959005833,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 0,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 0,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 4310.246666995226,
      "inferenceMs": 4304.8894169915,
      "pagesPerSecond": 0.2322940041277196,
      "tokensPerSecond": 0.2322940041277196,
      "outputTokens": 1,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 4310.246666995226,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 5629.381833001389,
      "inferenceMs": 5623.9088750007795,
      "pagesPerSecond": 0.17781226940663425,
      "tokensPerSecond": 9.068425739738347,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 5629.381833001389,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 5572.293749995879,
      "inferenceMs": 5566.85520800238,
      "pagesPerSecond": 0.17963463439151342,
      "tokensPerSecond": 9.161366353967185,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 5572.293749995879,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 5589.021207997575,
      "inferenceMs": 5583.2114160002675,
      "pagesPerSecond": 0.17910838861201242,
      "tokensPerSecond": 9.134527819212632,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 5589.021207997575,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 5578.438833996188,
      "inferenceMs": 5573.049500002526,
      "pagesPerSecond": 0.17943497541149542,
      "tokensPerSecond": 9.151183745986266,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 5578.438833996188,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 5578.071500000078,
      "inferenceMs": 5572.82766699791,
      "pagesPerSecond": 0.1794421180331782,
      "tokensPerSecond": 9.151548019692088,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 5578.071500000078,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 368.9181250083493,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 5576.827333003166,
      "inferenceMs": 5562.751667006523,
      "pagesPerSecond": 0.1797671476027131,
      "tokensPerSecond": 9.168124527738367,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 5945.745458011515,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 359.08891599683557,
      "placement": {
        "cpu/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 5497.921000001952,
      "inferenceMs": 5484.249167013331,
      "pagesPerSecond": 0.18234036593647154,
      "tokensPerSecond": 9.299358662760048,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "cpu"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 5857.009915998788,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 703.9827080006944,
      "placement": {
        "mps:0/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 3082.3674160055816,
      "inferenceMs": 3065.5722499941476,
      "pagesPerSecond": 0.32620337035015534,
      "tokensPerSecond": 16.636371887857923,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "mps:0"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "driverAllocatedMemoryBytes": 9401565184,
      "memoryMeasurementScope": "Post-inference Metal driver allocation, not peak",
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 3786.350124006276,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 2243.6030829994706,
      "inferenceMs": 2235.2628750086296,
      "pagesPerSecond": 0.44737467399494985,
      "tokensPerSecond": 20.131860329772742,
      "outputTokens": 45,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "mps:0"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "driverAllocatedMemoryBytes": 9401565184,
      "memoryMeasurementScope": "Post-inference Metal driver allocation, not peak",
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 2243.6030829994706,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 6642.127583007095,
      "inferenceMs": 6620.134625991341,
      "pagesPerSecond": 0.3021086598673977,
      "tokensPerSecond": 10.271694435491522,
      "outputTokens": 68,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "application/pdf",
      "activationPlacement": {
        "visionOutput": "mps:0"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "driverAllocatedMemoryBytes": 13965426688,
      "memoryMeasurementScope": "Post-inference Metal driver allocation, not peak",
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 6642.127583007095,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 0,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 0,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 1744.832915996085,
      "inferenceMs": 1735.5955000093672,
      "pagesPerSecond": 0.5761711182096305,
      "tokensPerSecond": 0.5761711182096305,
      "outputTokens": 1,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "mps:0"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "driverAllocatedMemoryBytes": 13965426688,
      "memoryMeasurementScope": "Post-inference Metal driver allocation, not peak",
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 1744.832915996085,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 2321.103750000475,
      "inferenceMs": 2313.114874996245,
      "pagesPerSecond": 0.4323174827197561,
      "tokensPerSecond": 22.048191618707563,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "mps:0"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "driverAllocatedMemoryBytes": 13973815296,
      "memoryMeasurementScope": "Post-inference Metal driver allocation, not peak",
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 2321.103750000475,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 2358.024834000389,
      "inferenceMs": 2348.791042008088,
      "pagesPerSecond": 0.4257509425551345,
      "tokensPerSecond": 21.71329807031186,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "mps:0"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "driverAllocatedMemoryBytes": 13973815296,
      "memoryMeasurementScope": "Post-inference Metal driver allocation, not peak",
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 2358.024834000389,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 2346.0327080101706,
      "inferenceMs": 2336.925124996924,
      "pagesPerSecond": 0.4279127257024618,
      "tokensPerSecond": 21.823549010825552,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "mps:0"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "driverAllocatedMemoryBytes": 13973815296,
      "memoryMeasurementScope": "Post-inference Metal driver allocation, not peak",
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 2346.0327080101706,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 2313.9295000000857,
      "inferenceMs": 2304.5750000019325,
      "pagesPerSecond": 0.4339194862389644,
      "tokensPerSecond": 22.129893798187187,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "mps:0"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "driverAllocatedMemoryBytes": 13973815296,
      "memoryMeasurementScope": "Post-inference Metal driver allocation, not peak",
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 2313.9295000000857,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 2330.975083998055,
      "inferenceMs": 2322.0267500000773,
      "pagesPerSecond": 0.43065826007386293,
      "tokensPerSecond": 21.963571263767008,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "mps:0"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "driverAllocatedMemoryBytes": 13973815296,
      "memoryMeasurementScope": "Post-inference Metal driver allocation, not peak",
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 2330.975083998055,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 653.638707997743,
      "placement": {
        "mps:0/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 2676.3502919930033,
      "inferenceMs": 2660.101790999761,
      "pagesPerSecond": 0.3759254639741303,
      "tokensPerSecond": 19.172198662680643,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "mps:0"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "driverAllocatedMemoryBytes": 9401565184,
      "memoryMeasurementScope": "Post-inference Metal driver allocation, not peak",
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 3329.9889999907464,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 710.6122920085909,
      "placement": {
        "mps:0/torch.float32": 608
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 2569.109625008423,
      "inferenceMs": 2552.866958998493,
      "pagesPerSecond": 0.39171645685457374,
      "tokensPerSecond": 19.97753929958326,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "mps:0"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "driverAllocatedMemoryBytes": 9401565184,
      "memoryMeasurementScope": "Post-inference Metal driver allocation, not peak",
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 3279.7219170170138,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp16",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 567.8210839978419,
      "placement": {
        "mps:0/torch.float16": 608
      },
      "parameterDtype": "torch.float16",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 2862.037832994247,
      "inferenceMs": 2787.506000007852,
      "pagesPerSecond": 0.3587436224342416,
      "tokensPerSecond": 18.295924744146323,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "mps:0"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "driverAllocatedMemoryBytes": 6220185600,
      "memoryMeasurementScope": "Post-inference Metal driver allocation, not peak",
      "device": "mps",
      "precision": "fp16",
      "cpuThreads": 4,
      "totalMs": 3429.858916992089,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "bf16",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "phraseRecall": 1,
      "minimumPhraseRecall": 0.8,
      "textTokenRecall": 1,
      "minimumTextTokenRecall": 0.9,
      "blankNoHallucination": true,
      "pageCoverage": true,
      "completeGeneration": true,
      "repetitive": false,
      "scope": "Independent authored content, not structured layout/table evaluation"
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "pdfiumVersion": "5.14.0",
      "loadMs": 618.5159159940667,
      "placement": {
        "mps:0/torch.bfloat16": 608
      },
      "parameterDtype": "torch.bfloat16",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "attentionImplementation": "sdpa",
      "accelerationDetails": "Vision/language parameters, inputs and vision activations on selected device; rendering, tokenizer and decoding on CPU",
      "durationMs": 2728.1076249928446,
      "inferenceMs": 2653.7129589996766,
      "pagesPerSecond": 0.3768305070857974,
      "tokensPerSecond": 19.218355861375667,
      "outputTokens": 51,
      "maxNewTokens": 512,
      "pdfScale": 1.5,
      "inputKind": "image",
      "activationPlacement": {
        "visionOutput": "mps:0"
      },
      "decoding": "greedy",
      "outputLimitReached": false,
      "allPagesReachedEos": true,
      "driverAllocatedMemoryBytes": 6220185600,
      "memoryMeasurementScope": "Post-inference Metal driver allocation, not peak",
      "device": "mps",
      "precision": "bf16",
      "cpuThreads": 4,
      "totalMs": 3346.6235409869114,
      "artifactRevision": "c5630abae1d940eafe0697512a0325494b02ab42",
      "fallback": false,
      "grade": "fast"
    }
  }
]

Something not quite right?

Tell us what happened with PaddleOCR-VL 1.6. 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.