Image-Text-to-Text
Transformers
Safetensors
Korean
English
qwen3_5
agent-safety
judge-model
behavioral-safety
lora
conversational
Instructions to use Tynapse/drift-sentry-4b-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Tynapse/drift-sentry-4b-v1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="Tynapse/drift-sentry-4b-v1") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("Tynapse/drift-sentry-4b-v1") model = AutoModelForMultimodalLM.from_pretrained("Tynapse/drift-sentry-4b-v1", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Tynapse/drift-sentry-4b-v1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Tynapse/drift-sentry-4b-v1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Tynapse/drift-sentry-4b-v1", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/Tynapse/drift-sentry-4b-v1
- SGLang
How to use Tynapse/drift-sentry-4b-v1 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "Tynapse/drift-sentry-4b-v1" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Tynapse/drift-sentry-4b-v1", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "Tynapse/drift-sentry-4b-v1" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Tynapse/drift-sentry-4b-v1", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }' - Docker Model Runner
How to use Tynapse/drift-sentry-4b-v1 with Docker Model Runner:
docker model run hf.co/Tynapse/drift-sentry-4b-v1
Download evaluation/label_consistency_audit.json from Tynapse/drift-sentry-4b-v1: direct link, hf CLI and curl.
- Browser
- Download file 4.32 kB
-
https://huggingface.co/Tynapse/drift-sentry-4b-v1/resolve/main/evaluation/label_consistency_audit.json
- Command line
-
hf download hf://Tynapse/drift-sentry-4b-v1/evaluation/label_consistency_audit.json
-
curl -L -o label_consistency_audit.json https://huggingface.co/Tynapse/drift-sentry-4b-v1/resolve/main/evaluation/label_consistency_audit.json
4.32 kB
| { | |
| "agreements": 146769, | |
| "cohen_kappa": 0.9781300729494184, | |
| "confusion": { | |
| "environment": { | |
| "environment": 17757, | |
| "goal": 0, | |
| "integration": 0, | |
| "memory": 0, | |
| "normal": 209, | |
| "reasoning": 2, | |
| "reward": 0 | |
| }, | |
| "goal": { | |
| "environment": 8, | |
| "goal": 26657, | |
| "integration": 0, | |
| "memory": 0, | |
| "normal": 289, | |
| "reasoning": 3, | |
| "reward": 0 | |
| }, | |
| "integration": { | |
| "environment": 2, | |
| "goal": 0, | |
| "integration": 21721, | |
| "memory": 0, | |
| "normal": 684, | |
| "reasoning": 4, | |
| "reward": 0 | |
| }, | |
| "memory": { | |
| "environment": 0, | |
| "goal": 0, | |
| "integration": 0, | |
| "memory": 14946, | |
| "normal": 45, | |
| "reasoning": 0, | |
| "reward": 0 | |
| }, | |
| "normal": { | |
| "environment": 37, | |
| "goal": 1, | |
| "integration": 7, | |
| "memory": 110, | |
| "normal": 14702, | |
| "reasoning": 137, | |
| "reward": 0 | |
| }, | |
| "reasoning": { | |
| "environment": 16, | |
| "goal": 0, | |
| "integration": 0, | |
| "memory": 0, | |
| "normal": 1072, | |
| "reasoning": 36147, | |
| "reward": 0 | |
| }, | |
| "reward": { | |
| "environment": 3, | |
| "goal": 0, | |
| "integration": 0, | |
| "memory": 0, | |
| "normal": 116, | |
| "reasoning": 1, | |
| "reward": 14839 | |
| } | |
| }, | |
| "disagreements": 2746, | |
| "distribution_audit": { | |
| "environment": { | |
| "deviation_percentage_points": 0.0014135783535813573, | |
| "observed_percent": 12.001413578353581, | |
| "planned_percent": 12.0, | |
| "rows": 17999 | |
| }, | |
| "goal": { | |
| "deviation_percentage_points": 0.0011201941669902737, | |
| "observed_percent": 18.00112019416699, | |
| "planned_percent": 18.0, | |
| "rows": 26997 | |
| }, | |
| "integration": { | |
| "deviation_percentage_points": 0.002600450744795424, | |
| "observed_percent": 15.002600450744795, | |
| "planned_percent": 15.0, | |
| "rows": 22500 | |
| }, | |
| "memory": { | |
| "deviation_percentage_points": 0.0017336338298630238, | |
| "observed_percent": 10.001733633829863, | |
| "planned_percent": 10.0, | |
| "rows": 15000 | |
| }, | |
| "normal": { | |
| "deviation_percentage_points": 0.0017336338298630238, | |
| "observed_percent": 10.001733633829863, | |
| "planned_percent": 10.0, | |
| "rows": 15000 | |
| }, | |
| "reasoning": { | |
| "deviation_percentage_points": -0.009001560270448294, | |
| "observed_percent": 24.99099843972955, | |
| "planned_percent": 25.0, | |
| "rows": 37480 | |
| }, | |
| "reward": { | |
| "deviation_percentage_points": 0.00040006934535341543, | |
| "observed_percent": 10.000400069345353, | |
| "planned_percent": 10.0, | |
| "rows": 14998 | |
| } | |
| }, | |
| "exclusion_counts": { | |
| "missing_or_invalid_raw_teacher_tier": 459 | |
| }, | |
| "expected_agreement": 0.16021438084377937, | |
| "inputs": [ | |
| "semantic_150k/train.jsonl", | |
| "semantic_150k/val.jsonl", | |
| "semantic_150k/test.jsonl" | |
| ], | |
| "labels": [ | |
| "goal", | |
| "reasoning", | |
| "environment", | |
| "integration", | |
| "memory", | |
| "reward", | |
| "normal" | |
| ], | |
| "maximum_absolute_distribution_deviation_percentage_points": 0.009001560270448294, | |
| "method": "Cohen's kappa between the independently assigned target_tier and the tier in metadata.raw_teacher_text.label, before any target-tier overwrite", | |
| "observed_agreement": 0.981633949770926, | |
| "planned_distribution_percent": { | |
| "environment": 12.0, | |
| "goal": 18.0, | |
| "integration": 15.0, | |
| "memory": 10.0, | |
| "normal": 10.0, | |
| "reasoning": 25.0, | |
| "reward": 10.0 | |
| }, | |
| "raw_teacher_counts_valid_rows": { | |
| "environment": 17823, | |
| "goal": 26658, | |
| "integration": 21728, | |
| "memory": 15056, | |
| "normal": 17117, | |
| "reasoning": 36294, | |
| "reward": 14839 | |
| }, | |
| "rows_excluded_from_kappa": 459, | |
| "rows_total": 149974, | |
| "rows_valid_for_kappa": 149515, | |
| "schema_version": "judge6.label-consistency-audit.v1", | |
| "target_counts_all_rows": { | |
| "environment": 17999, | |
| "goal": 26997, | |
| "integration": 22500, | |
| "memory": 15000, | |
| "normal": 15000, | |
| "reasoning": 37480, | |
| "reward": 14998 | |
| }, | |
| "target_counts_valid_rows": { | |
| "environment": 17968, | |
| "goal": 26957, | |
| "integration": 22411, | |
| "memory": 14991, | |
| "normal": 14994, | |
| "reasoning": 37235, | |
| "reward": 14959 | |
| }, | |
| "redaction": { | |
| "removed": [ | |
| "absolute filesystem paths" | |
| ], | |
| "internal_evidence_retained": true | |
| } | |
| } | |