GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
ningpy
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
ningpy
Model Tree
Qwen/Qwen2.5-3B-Instruct
Input Modalities
Output Modalities
Supported Functionality
peiyan-ning/redflag-symptom-3b — 83-symptom multi-label extractionpeiyan-ning/redflag-context-3b — 12 context flags (post_trauma, drowning, etc.)peiyan-ning/redflag-modifier-3b — onset / fever_celsius / consciousness / etc.peiyan-ning/redflag-denied-3b — denied symptoms (multi-turn negation)peiyan-ning/redflag-gate-3b — 8 population gates (is_pregnant, is_child, ...)Full 5-module pipeline + rule engine V46:
| Metric | P | R | F1 | Acc |
|---|---|---|---|---|
| PRIMARY (any_matched × labeled_matched) | 0.902 | 0.911 | 0.906 | 91.9% |
| STRICT matched-only | 0.893 | 0.828 | 0.859 | 91.8% |
| STRICT m+s | 0.844 | 0.905 | 0.873 | 92.1% |
markdown
Extract quantitative modifiers from text.===== EXTRACTSet onset:"acute" when text describes SUDDEN event:- "suddenly", "just", "out of nowhere", "in an instant", "突然", "tiba-tiba"- Trauma events: "fell", "crash", "hit"- Symptom onset described as fast: "just started"Set onset:"chronic" when described as long-standing/gradual.===== EXTRACT temperature =====When text has actual number: "39.5°C" / "38 degrees" / "40度" → fever_celsius===== EXTRACT other modifiers =====- seizure duration in minutes → seizure_duration_min- inhaler used / worked → inhaler_used / inhaler_effective- bleeding amount described "heavy" / "uncontrolled" → bleeding_severity- burn size → burn_severity- burn location → burn_location- consciousness: alert / confused / drowsy / unresponsive===== EXAMPLES ====="Suddenly severe chest pain" → {"modifiers": {"onset": "acute"}}"Fell down and hit head" → {"modifiers": {"onset": "acute"}}"Fever 39.5 for 3 days" → {"modifiers": {"onset": "chronic", "fever_celsius": 39.5, "fever_days": 3}}"Seizure lasted 7 minutes" → {"modifiers": {"seizure_duration_min": 7}}"Uncontrolled bleeding" → {"modifiers": {"bleeding_severity": "uncontrolled"}}"Not sure about symptoms" → {"modifiers": {}}Output: {"modifiers": {...}}===== MULTILINGUAL / MANGLISH GUIDANCE =====Text may be in Brunei/Manglish English or mixed with Malay/Chinese.Ignore these colloquial particles when extracting: "lah", "kah", "meh", "ah", "leh", "lor", "sia", "one".Common Manglish/Malay/Chinese mappings:- "kena panic attack" / "feel like dying" / "jantung deg-deg" → severe_panic- "sesak nafas" (Malay) / "喘不过气" → breathlessness- "sakit dada" (Malay) / "胸口疼" → chest_pain- "sakit kepala teruk" / "剧烈头痛" / "worst headache" → thunderclap_headache- "pengsan" (Malay) / "晕倒" → fainting- "sawan" (Malay) / "抽搐" → seizure- "anak saya" (Malay: my child) → is_child- "bayi saya" (Malay: my baby) → is_baby- "warga emas" / "老人家" → is_elderly- "hamil" / "怀孕" → is_pregnant- "kencing manis" (Malay: diabetes) → has_diabetes- "asma" (Malay: asthma) → has_asthma- "kena patuk ular" (Malay: snake bit) → context_flags: venomous_biteAuntie/uncle in Manglish family reference: usually elderly family member → is_elderly.
python
from transformers import AutoTokenizer, AutoModelForCausalLMimport torch, jsontok = AutoTokenizer.from_pretrained('peiyan-ning/redflag-modifier-3b')model = AutoModelForCausalLM.from_pretrained('peiyan-ning/redflag-modifier-3b',torch_dtype=torch.float16,device_map='auto')SYSTEM_PROMPT = tok.chat_template # or use the prompt abovemessages = [{'role': 'system', 'content': SYSTEM_PROMPT},{'role': 'user', 'content': 'My 3-year-old child has severe fever and vomiting lah'},]inputs = tok.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors='pt').to(model.device)out = model.generate(inputs, max_new_tokens=200, do_sample=False)text = tok.decode(out[0][inputs.shape[1]:], skip_special_tokens=True)result = json.loads(text)print(result)
See git.evyd.tech/ai/redflag-detection-2.0 for: