OpenMed-mLiteClinical-IrishPPSN-135M-v1-q8

OpenMed-mLiteClinical-IrishPPSN-135M-v1-q8

OpenMed-mLiteClinical-IrishPPSN-135M-v1-q8 is a CPU-oriented dynamic int8 export of temsa/OpenMed-mLiteClinical-IrishPPSN-135M-v1.

Use this release when you want a smaller, faster CPU deployment and can accept some accuracy loss relative to the full checkpoint.

What This Release Is

• A dynamic int8 torch export stored as model_quantized.pt
• Derived from temsa/OpenMed-mLiteClinical-IrishPPSN-135M-v1
• Intended for PPSN masking with the bundled word_aligned decoder
• Tuned for CPU inference, not GPU inference

Important Limitation

This is not a standard AutoModelForTokenClassification.from_pretrained(...) package. Use the bundled inference_word_aligned.py loader, which knows how to download and load model_quantized.pt.

Recommended Inference

python3 inference_word_aligned.py --ppsn-min-score 0.7 --text "My PPSN is 1234567TW and I need help with my housing grant." --json

To load directly from the Hub:

python3 inference_word_aligned.py --model temsa/OpenMed-mLiteClinical-IrishPPSN-135M-v1-q8 --ppsn-min-score 0.7 --text "My PPSN is 1234567TW and I need help with my housing grant." --json

Benchmark Summary

Measured on the multilingual PPSN suite spanning:

• gov data
• citizen-to-government chat
• HSE medical text
• English, Irish Gaelic, and additional European / Ukrainian / Russian / Chinese / Japanese examples

Tradeoff

• Q8 is much faster on CPU than the full checkpoint
• Q8 still clearly outperforms the original base model on PPSN detection
• Q8 is less accurate than the full checkpoint and needs a higher PPSN threshold

Included Files

• Quantized model:
  • model_quantized.pt
  • quantization.json
• Tokenizer / config:
  • config.json
  • tokenizer.json
  • tokenizer_config.json
  • special_tokens_map.json
  • vocab.txt
  • label_meta.json
• Inference / QA:
  • inference_word_aligned.py
  • qa_config.json
  • pyproject.toml
• Benchmark artifacts:
  • eval/

License

This quantized derivative is distributed under Apache-2.0, consistent with the canonical full model and the upstream base model. See NOTICE for attribution.

<!-- portfolio-comparison:start -->

Portfolio Comparison

Updated: 2026-03-16.

Use this section for the fastest public comparison across the temsa PII masking portfolio.

• The first core table only includes public checkpoints that ship both comparable q8 accuracy and q8 CPU throughput.
• The first PPSN table only includes public artifacts that ship comparable PPSN accuracy and CPU throughput.
• Missing cells in the archive tables mean the older release did not ship that metric in its public bundle.
• DiffMask rows use the reconciled clean_single_pass harness that matches the deployed runtime.
• GlobalPointer rows use the public raw-only span-matrix release bundle and its packaged q8 ONNX artifact.
• The same content is shipped as PORTFOLIO_COMPARISON.md inside each public model repo.

Irish Core PII: Comparable Public Checkpoints

Irish Core PII: Other Public Checkpoints

Finance-boundary q8 F1 is 1.0000 for OpenMed-mLiteClinical-IrishCorePII-135M-v2-rc6, OpenMed-mLiteClinical-IrishCorePII-135M-v2-rc7, OpenMed-mLiteClinical-IrishCorePII-135M-v2-rc8, and all public IrishCore-DiffMask releases from rc1 to rc6. OpenMed-mLiteClinical-IrishCorePII-135M-v2-rc5 ships 0.8750 on that public q8 suite.

PPSN-Only: Comparable Public Artifacts

PPSN-Only: Historical Public Checkpoints

If you need the strongest current raw-only Irish core model, start with IrishCore-GlobalPointer-135M-v1-rc4. If you need the fastest CPU-first raw-only line, compare it against IrishCore-DiffMask-135M-v1-rc6. If you need a PPSN-only artifact, compare the canonical fp32, fp16, and q8 variants of OpenMed-mLiteClinical-IrishPPSN-135M-v1 directly in the table above.

<!-- portfolio-comparison:end -->