Image-Text-to-Text
Transformers
Safetensors
qwen2_5_vl
qwen3_vl
vision-language
visual-token-pruning
visionzip
scopd
Instructions to use enmingzhangzz/SCOPD with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use enmingzhangzz/SCOPD with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="enmingzhangzz/SCOPD")# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("enmingzhangzz/SCOPD", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use enmingzhangzz/SCOPD with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "enmingzhangzz/SCOPD" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "enmingzhangzz/SCOPD", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/enmingzhangzz/SCOPD
- SGLang
How to use enmingzhangzz/SCOPD 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 "enmingzhangzz/SCOPD" \ --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": "enmingzhangzz/SCOPD", "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 "enmingzhangzz/SCOPD" \ --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": "enmingzhangzz/SCOPD", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use enmingzhangzz/SCOPD with Docker Model Runner:
docker model run hf.co/enmingzhangzz/SCOPD
Download models.json from enmingzhangzz/SCOPD: direct link, hf CLI and curl.
- Browser
- Download file 10.8 kB
-
https://huggingface.co/enmingzhangzz/SCOPD/resolve/main/models.json
- Command line
-
hf download hf://enmingzhangzz/SCOPD/models.json
-
curl -L -o models.json https://huggingface.co/enmingzhangzz/SCOPD/resolve/main/models.json
10.8 kB
| { | |
| "base_model": "Qwen/Qwen2.5-VL-7B-Instruct", | |
| "base_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5", | |
| "models": [ | |
| { | |
| "id": "scopd", | |
| "label": "SCOPD", | |
| "objective": "forward KL: KL(teacher || student)", | |
| "selection": "all response tokens", | |
| "adapter_sha256": "0f882974e3e8565c7d9a99494e11289736c384e0d60fb9ba9e20ed5417a7b6a9", | |
| "source_revision": "c7182c3867f9f8819b23af4bd57ba785c4daf5c8", | |
| "base_model": "Qwen/Qwen2.5-VL-7B-Instruct", | |
| "base_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5", | |
| "merged": false, | |
| "format": "peft_lora", | |
| "dtype": "float32", | |
| "inference_dtype": "bfloat16", | |
| "questions_processed": 10240, | |
| "path": "models/scopd", | |
| "tensor_count": 392, | |
| "parameter_count": 40370176, | |
| "weight_bytes": 161480704, | |
| "lora_rank": 16, | |
| "lora_alpha": 32, | |
| "lora_dropout": 0.0 | |
| }, | |
| { | |
| "id": "scopd-plus", | |
| "label": "SCOPD+", | |
| "objective": "forward KL: KL(teacher || student)", | |
| "selection": "top 10% budget-JSD response tokens", | |
| "selection_fraction": 0.1, | |
| "budget_delta": 0.01, | |
| "adapter_sha256": "04c4590b14355f3767ebaa63a435f42def65fa83d386624b6ec1a75a23315e8d", | |
| "source_revision": "863f3f0097db7eb4b9bca1a14d862edf138d407f", | |
| "base_model": "Qwen/Qwen2.5-VL-7B-Instruct", | |
| "base_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5", | |
| "merged": false, | |
| "format": "peft_lora", | |
| "dtype": "float32", | |
| "inference_dtype": "bfloat16", | |
| "questions_processed": 10240, | |
| "path": "models/scopd-plus", | |
| "tensor_count": 392, | |
| "parameter_count": 40370176, | |
| "weight_bytes": 161480704, | |
| "lora_rank": 16, | |
| "lora_alpha": 32, | |
| "lora_dropout": 0.0 | |
| }, | |
| { | |
| "id": "scopd-plus-top20", | |
| "label": "SCOPD+ Top20", | |
| "objective": "forward KL: KL(teacher || student)", | |
| "selection": "top 20% budget-JSD response tokens", | |
| "selection_fraction": 0.2, | |
| "budget_delta": 0.01, | |
| "adapter_sha256": "4f70e51e1c9cbfe953f7277786c82a32c6b754919bb6c8307fa2012f1ab8ec57", | |
| "source_revision": "52b62edc3c51c310a75278ed53ca0187ed333a82", | |
| "base_model": "Qwen/Qwen2.5-VL-7B-Instruct", | |
| "base_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5", | |
| "merged": false, | |
| "format": "peft_lora", | |
| "dtype": "float32", | |
| "inference_dtype": "bfloat16", | |
| "questions_processed": 10240, | |
| "path": "models/scopd-plus-top20", | |
| "tensor_count": 392, | |
| "parameter_count": 40370176, | |
| "weight_bytes": 161480704, | |
| "lora_rank": 16, | |
| "lora_alpha": 32, | |
| "lora_dropout": 0.0 | |
| }, | |
| { | |
| "id": "scopd-plus-top40", | |
| "label": "SCOPD+ Top40", | |
| "objective": "forward KL: KL(teacher || student)", | |
| "selection": "top 40% budget-JSD response tokens", | |
| "selection_fraction": 0.4, | |
| "budget_delta": 0.01, | |
| "adapter_sha256": "3d8791563dd4a732b3f9fe072a6f2588faa0bc192de131794d4914c60a658497", | |
| "base_model": "Qwen/Qwen2.5-VL-7B-Instruct", | |
| "base_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5", | |
| "merged": false, | |
| "format": "peft_lora", | |
| "dtype": "float32", | |
| "inference_dtype": "bfloat16", | |
| "questions_processed": 10240, | |
| "path": "models/scopd-plus-top40", | |
| "tensor_count": 392, | |
| "parameter_count": 40370176, | |
| "weight_bytes": 161480704, | |
| "lora_rank": 16, | |
| "lora_alpha": 32, | |
| "lora_dropout": 0.0 | |
| }, | |
| { | |
| "id": "scopd-plus-top60", | |
| "label": "SCOPD+ Top60", | |
| "objective": "forward KL: KL(teacher || student)", | |
| "selection": "top 60% budget-JSD response tokens", | |
| "selection_fraction": 0.6, | |
| "budget_delta": 0.01, | |
| "adapter_sha256": "c02297cfe2fcd3ab01d769d0aca53cb4a3bb43dba6892895ecc282f23b053b3a", | |
| "base_model": "Qwen/Qwen2.5-VL-7B-Instruct", | |
| "base_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5", | |
| "merged": false, | |
| "format": "peft_lora", | |
| "dtype": "float32", | |
| "inference_dtype": "bfloat16", | |
| "questions_processed": 10240, | |
| "path": "models/scopd-plus-top60", | |
| "tensor_count": 392, | |
| "parameter_count": 40370176, | |
| "weight_bytes": 161480704, | |
| "lora_rank": 16, | |
| "lora_alpha": 32, | |
| "lora_dropout": 0.0 | |
| }, | |
| { | |
| "id": "sft", | |
| "label": "SFT", | |
| "objective": "supervised fine-tuning", | |
| "selection": "supervised response tokens", | |
| "adapter_sha256": "9463250e17a196055c2f54e205587c7e727c3b5b6b7a30eb3158c5967ccb9f8e", | |
| "source_revision": "0c0d37ac6d9cbc0243dbf431d426e16931e5c79f", | |
| "base_model": "Qwen/Qwen2.5-VL-7B-Instruct", | |
| "base_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5", | |
| "merged": false, | |
| "format": "peft_lora", | |
| "dtype": "float32", | |
| "inference_dtype": "bfloat16", | |
| "questions_processed": 10240, | |
| "path": "models/sft", | |
| "tensor_count": 392, | |
| "parameter_count": 40370176, | |
| "weight_bytes": 161480704, | |
| "lora_rank": 16, | |
| "lora_alpha": 32, | |
| "lora_dropout": 0.0 | |
| }, | |
| { | |
| "id": "epic", | |
| "label": "EPIC", | |
| "objective": "official EPIC configuration", | |
| "selection": "EPIC response supervision", | |
| "adapter_sha256": "86acec7d328a0a895c695dbca7582f38ef0ddda40817f000e82830c50c38555b", | |
| "source_revision": "0d0fe8bd4f4a1b280937d97a9c2839c814cf6a9d", | |
| "base_model": "Qwen/Qwen2.5-VL-7B-Instruct", | |
| "base_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5", | |
| "merged": false, | |
| "format": "peft_lora", | |
| "dtype": "float32", | |
| "inference_dtype": "bfloat16", | |
| "questions_processed": 10240, | |
| "path": "models/epic", | |
| "tensor_count": 392, | |
| "parameter_count": 40370176, | |
| "weight_bytes": 161480704, | |
| "lora_rank": 16, | |
| "lora_alpha": 32, | |
| "lora_dropout": 0.0 | |
| }, | |
| { | |
| "id": "grpo", | |
| "label": "GRPO", | |
| "objective": "group-relative policy optimization", | |
| "selection": "reward-based policy objective", | |
| "adapter_sha256": "c38704d69eb2e1ac0164d9ec1be644d42449ff80eb130175855550c94745bcf4", | |
| "optimizer_updates": 2560, | |
| "base_model": "Qwen/Qwen2.5-VL-7B-Instruct", | |
| "base_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5", | |
| "merged": false, | |
| "format": "peft_lora", | |
| "dtype": "float32", | |
| "inference_dtype": "bfloat16", | |
| "questions_processed": 10240, | |
| "path": "models/grpo", | |
| "tensor_count": 392, | |
| "parameter_count": 40370176, | |
| "weight_bytes": 161480704, | |
| "lora_rank": 16, | |
| "lora_alpha": 32, | |
| "lora_dropout": 0.0 | |
| }, | |
| { | |
| "id": "scopd-offpolicy", | |
| "label": "SCOPD Off-Policy", | |
| "objective": "forward KL on teacher-generated responses", | |
| "selection": "all response tokens", | |
| "adapter_sha256": "dbf02470f09a41bfe8bad808a17f96ba892c582e7510d12daf2df0e1047b9281", | |
| "base_model": "Qwen/Qwen2.5-VL-7B-Instruct", | |
| "base_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5", | |
| "merged": false, | |
| "format": "peft_lora", | |
| "dtype": "float32", | |
| "inference_dtype": "bfloat16", | |
| "questions_processed": 10240, | |
| "path": "models/scopd-offpolicy", | |
| "tensor_count": 392, | |
| "parameter_count": 40370176, | |
| "weight_bytes": 161480704, | |
| "lora_rank": 16, | |
| "lora_alpha": 32, | |
| "lora_dropout": 0.0 | |
| }, | |
| { | |
| "id": "scopd-plus-reverse-kl", | |
| "label": "SCOPD+ Reverse KL", | |
| "objective": "reverse KL: KL(student || teacher)", | |
| "selection": "top 10% budget-JSD response tokens", | |
| "selection_fraction": 0.1, | |
| "budget_delta": 0.01, | |
| "adapter_sha256": "1a7c4004928c80f85a66d216c6d17b17005fe89740da946d7597225775b00baa", | |
| "source_revision": "dc2490eaf9b8b35b7e5553080dd985d0ab6a86a3", | |
| "base_model": "Qwen/Qwen2.5-VL-7B-Instruct", | |
| "base_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5", | |
| "merged": false, | |
| "format": "peft_lora", | |
| "dtype": "float32", | |
| "inference_dtype": "bfloat16", | |
| "questions_processed": 10240, | |
| "path": "models/scopd-plus-reverse-kl", | |
| "tensor_count": 392, | |
| "parameter_count": 40370176, | |
| "weight_bytes": 161480704, | |
| "lora_rank": 16, | |
| "lora_alpha": 32, | |
| "lora_dropout": 0.0 | |
| }, | |
| { | |
| "id": "scopd-plus-jsd", | |
| "label": "SCOPD+ JSD", | |
| "objective": "symmetric JSD(teacher, student)", | |
| "selection": "top 10% budget-JSD response tokens", | |
| "selection_fraction": 0.1, | |
| "budget_delta": 0.01, | |
| "adapter_sha256": "ea78a9ae00822a76a55824846b385662d9baae5998dcffa928a45be9f177c954", | |
| "base_model": "Qwen/Qwen2.5-VL-7B-Instruct", | |
| "base_revision": "cc594898137f460bfe9f0759e9844b3ce807cfb5", | |
| "merged": false, | |
| "format": "peft_lora", | |
| "dtype": "float32", | |
| "inference_dtype": "bfloat16", | |
| "questions_processed": 10240, | |
| "path": "models/scopd-plus-jsd", | |
| "tensor_count": 392, | |
| "parameter_count": 40370176, | |
| "weight_bytes": 161480704, | |
| "lora_rank": 16, | |
| "lora_alpha": 32, | |
| "lora_dropout": 0.0 | |
| }, | |
| { | |
| "id": "qwen3-vl-4b-scopd", | |
| "label": "Qwen3-VL-4B SCOPD", | |
| "objective": "forward KL: KL(teacher || student)", | |
| "selection": "all response tokens", | |
| "adapter_sha256": "6aec6fba5176359d4d1a3d906d3722a51b9f7f21c79d5d5989120b075915102f", | |
| "optimizer_updates": 320, | |
| "base_model": "Qwen/Qwen3-VL-4B-Instruct", | |
| "base_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17", | |
| "merged": false, | |
| "format": "peft_lora", | |
| "dtype": "float32", | |
| "inference_dtype": "bfloat16", | |
| "questions_processed": 10240, | |
| "path": "models/qwen3-vl-4b-scopd", | |
| "tensor_count": 504, | |
| "parameter_count": 33030144, | |
| "weight_bytes": 132120576, | |
| "lora_rank": 16, | |
| "lora_alpha": 32, | |
| "lora_dropout": 0.0 | |
| }, | |
| { | |
| "id": "qwen3-vl-4b-scopd-plus", | |
| "label": "Qwen3-VL-4B SCOPD+", | |
| "objective": "forward KL: KL(teacher || student)", | |
| "selection": "top 10% budget-JSD response tokens", | |
| "selection_fraction": 0.1, | |
| "budget_delta": 0.01, | |
| "adapter_sha256": "febccb411829f5c314d421038c1cd89dcc543df074978778771504ab1dc91a0e", | |
| "optimizer_updates": 320, | |
| "base_model": "Qwen/Qwen3-VL-4B-Instruct", | |
| "base_revision": "ebb281ec70b05090aa6165b016eac8ec08e71b17", | |
| "merged": false, | |
| "format": "peft_lora", | |
| "dtype": "float32", | |
| "inference_dtype": "bfloat16", | |
| "questions_processed": 10240, | |
| "path": "models/qwen3-vl-4b-scopd-plus", | |
| "tensor_count": 504, | |
| "parameter_count": 33030144, | |
| "weight_bytes": 132120576, | |
| "lora_rank": 16, | |
| "lora_alpha": 32, | |
| "lora_dropout": 0.0 | |
| } | |
| ] | |
| } | |