Text Classification
Transformers
Safetensors
qwen3
content-safety
prompt-injection
multi-label-classification
safety
multilingual
hikmaai
Instructions to use HikmaAI/HikmaGuard with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use HikmaAI/HikmaGuard with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="HikmaAI/HikmaGuard")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("HikmaAI/HikmaGuard") model = AutoModelForSequenceClassification.from_pretrained("HikmaAI/HikmaGuard", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download thresholds.json from HikmaAI/HikmaGuard: direct link, hf CLI and curl.
- Browser
- Download file 894 Bytes
-
https://huggingface.co/HikmaAI/HikmaGuard/resolve/main/thresholds.json
- Command line
-
hf download hf://HikmaAI/HikmaGuard/thresholds.json
-
curl -L -o thresholds.json https://huggingface.co/HikmaAI/HikmaGuard/resolve/main/thresholds.json
894 Bytes
| { | |
| "schema_version": 1, | |
| "taxonomy_version": "unified-guard-v1", | |
| "policy_version": "unified-guard-v1", | |
| "classifier_name": "unified_guard", | |
| "model_config_sha256": "f1b8025a331a052b2117acb5c0ec3ccbded6904655c41b1ea7365de78277e138", | |
| "model_artifact_id": "d21042f88a814cc19011421f0cf972be", | |
| "heads": { | |
| "hate": { | |
| "threshold": 0.15102779865264893 | |
| }, | |
| "abuse": { | |
| "threshold": 0.32167303562164307 | |
| }, | |
| "violence": { | |
| "threshold": 0.5765419602394104 | |
| }, | |
| "self_harm": { | |
| "threshold": 0.14318770170211792 | |
| }, | |
| "sexual_harm": { | |
| "threshold": 0.1919327825307846 | |
| }, | |
| "misconduct": { | |
| "threshold": 0.7745833992958069 | |
| }, | |
| "dangerous_capability": { | |
| "threshold": 0.7549149990081787 | |
| }, | |
| "prompt_injection": { | |
| "threshold": 0.98 | |
| }, | |
| "indirect_prompt_injection": { | |
| "threshold": 0.992 | |
| } | |
| } | |
| } | |