Image-Text-to-Text
Transformers
Safetensors
English
pathology
computational-pathology
digital-pathology
histopathology
whole-slide-image
vision-language-model
report-generation
synoptic-report
case-level
conch
qwen2.5
Eval Results (legacy)
Instructions to use AtlasAnalyticsLab/PathoSynVLM with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AtlasAnalyticsLab/PathoSynVLM with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="AtlasAnalyticsLab/PathoSynVLM")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("AtlasAnalyticsLab/PathoSynVLM", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use AtlasAnalyticsLab/PathoSynVLM with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "AtlasAnalyticsLab/PathoSynVLM" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "AtlasAnalyticsLab/PathoSynVLM", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/AtlasAnalyticsLab/PathoSynVLM
- SGLang
How to use AtlasAnalyticsLab/PathoSynVLM 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 "AtlasAnalyticsLab/PathoSynVLM" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "AtlasAnalyticsLab/PathoSynVLM", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'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 "AtlasAnalyticsLab/PathoSynVLM" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "AtlasAnalyticsLab/PathoSynVLM", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use AtlasAnalyticsLab/PathoSynVLM with Docker Model Runner:
docker model run hf.co/AtlasAnalyticsLab/PathoSynVLM
| { | |
| "best_step": 30400, | |
| "best_epoch": 7, | |
| "best_source": "step_eval", | |
| "best_val_loss": 1.0108920872211455, | |
| "metrics": { | |
| "val_loss": 1.0108920872211455, | |
| "val_n_samples": 30, | |
| "val/vision_tokens_total": 34833, | |
| "val/text_tokens_total": 9942, | |
| "val/target_tokens_total": 3305, | |
| "val/wsi_count_total": 188, | |
| "val/avg_vision_tokens_per_sample": 1161.1, | |
| "val/avg_text_tokens_per_sample": 331.4, | |
| "val/avg_target_tokens_per_sample": 110.16666666666667, | |
| "val/avg_wsi_per_sample": 6.266666666666667, | |
| "val_bertscore_f1": 0.3017574870338043, | |
| "val/field_eval_n": 30, | |
| "val_diagnosis_relaxed_match_rate": 0.3333333333333333, | |
| "val_diagnosis_exact_match_rate": 0.16666666666666666, | |
| "val_certainty_match_rate": 0.9, | |
| "val_certainty_exact_match_rate": 0.9, | |
| "val_conclusion_exact_match_rate": 0.0, | |
| "val_conclusion_rougeL": 0.24946812913260996, | |
| "val_conclusion_meteor": 0.1988453811627603, | |
| "val_conclusion_bleu4": 5.251211118089842, | |
| "val/pred_format_score_0_count": 0, | |
| "val/ref_format_score_0_count": 0, | |
| "val/pred_format_score_1_count": 0, | |
| "val/ref_format_score_1_count": 0, | |
| "val/pred_format_score_2_count": 0, | |
| "val/ref_format_score_2_count": 0, | |
| "val/pred_format_score_3_count": 30, | |
| "val/ref_format_score_3_count": 30, | |
| "val/pred_diagnosis_present_rate": 1.0, | |
| "val/pred_certainty_present_rate": 1.0, | |
| "val/pred_conclusion_present_rate": 1.0 | |
| } | |
| } |