Open weights
The models border runs
The 11 published models this library reaches: the PII tagger and 10 classifiers. Each carries the export gate that let it ship, and the languages it was actually trained in.
11models, all Apache-2.0
26languages, every model evaluated in each
13–146 msp95 per model-backed detector, 87 tokens on one CPU thread
13 of 13
| Select | Model | Wired to | Languages | Base | Artifact | Size | Export gate |
|---|---|---|---|---|---|---|---|
| piiguardtoken-classificationOpenNERXLM-RoBERTa base · 555 MB | piioutput_leakage | 26 of 26 | XLM-RoBERTa base | ONNX FP16ONNX | 555 MB | 300/300 unchanged, 0 chars uncovered | |
| toxicitymulti-label-classificationXLM-RoBERTa base · 535 MB | toxicity | 26 of 26 | XLM-RoBERTa base | ONNX INT8ONNX | 535 MB | 0/300 decisions moved | |
| nsfwmulti-label-classificationXLM-RoBERTa base · 535 MB | nsfw | 26 of 26 | XLM-RoBERTa base | ONNX INT8ONNX | 535 MB | 0/300 decisions moved | |
| biasmulti-label-classificationXLM-RoBERTa base · 535 MB | bias | 26 of 26 | XLM-RoBERTa base | ONNX INT8ONNX | 535 MB | 0/300 decisions moved | |
| gibberishtext-classificationXLM-RoBERTa base · 535 MB | gibberish | 26 of 26 | XLM-RoBERTa base | ONNX INT8ONNX | 535 MB | 0/300 decisions moved | |
| politenesstext-classificationXLM-RoBERTa base · 535 MB | politeness | 26 of 26 | XLM-RoBERTa base | ONNX INT8ONNX | 535 MB | 0/300 decisions moved | |
| injectiontext-classificationXLM-RoBERTa base · 535 MB | injection | 26 of 26 | XLM-RoBERTa base | ONNX INT8ONNX | 535 MB | 0/300 decisions moved | |
| regulated_advicemulti-label-classificationXLM-RoBERTa base · 535 MB | regulated_advice | 26 of 26 | XLM-RoBERTa base | ONNX INT8ONNX | 535 MB | 0/300 decisions moved | |
| groundednesstext-classificationXLM-RoBERTa base · 556 MB | groundedness | 26 of 26 | XLM-RoBERTa base | ONNX FP16ONNX | 556 MB | 0/300 decisions moved | |
| topic_scope_v3text-classificationXLM-RoBERTa large + decision head · 1571 MB | topic_scope | 26 of 26 | XLM-RoBERTa large + decision head | ONNX INT8 | 1571 MB | 0/300 decisions moved | |
| topic_scope_v2text-classificationXLM-RoBERTa base + decision head · 592 MB | supersededBacked topic_scope from 2026-09-30 to 2026-10-01. Replaced by topic-scope-v3, the same decision head on XLM-RoBERTa large. It remains published on the Hub. | 26 of 26 | XLM-RoBERTa base + decision head | ONNX INT8 | 592 MB | 0/300 decisions moved | |
| topic_scopesentence-similarityXLM-RoBERTa base · 533 MB | supersededBacked topic_scope until 2026-09-30. Replaced by topic-scope-v2: a bi-encoder scores similarity to each node, so it cannot read a node described by exclusion or answer that a message is about none of them. | 26 of 26 | XLM-RoBERTa base | ONNX INT8ONNX | 533 MB | 2/200 nodes moved, cos 0.9993 | |
| moderationmulti-label-classificationXLM-RoBERTa base · 364 MB | moderation | 26 of 26 | XLM-RoBERTa base | ONNX INT8+FP16ONNX | 364 MB | 0/300 decisions moved |