Instructions to use Yuan-Che/OpenECADv2-SigLIP-0.89B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Yuan-Che/OpenECADv2-SigLIP-0.89B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Yuan-Che/OpenECADv2-SigLIP-0.89B", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("Yuan-Che/OpenECADv2-SigLIP-0.89B", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Yuan-Che/OpenECADv2-SigLIP-0.89B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Yuan-Che/OpenECADv2-SigLIP-0.89B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Yuan-Che/OpenECADv2-SigLIP-0.89B", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Yuan-Che/OpenECADv2-SigLIP-0.89B
- SGLang
How to use Yuan-Che/OpenECADv2-SigLIP-0.89B 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 "Yuan-Che/OpenECADv2-SigLIP-0.89B" \ --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": "Yuan-Che/OpenECADv2-SigLIP-0.89B", "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 "Yuan-Che/OpenECADv2-SigLIP-0.89B" \ --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": "Yuan-Che/OpenECADv2-SigLIP-0.89B", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Yuan-Che/OpenECADv2-SigLIP-0.89B with Docker Model Runner:
docker model run hf.co/Yuan-Che/OpenECADv2-SigLIP-0.89B
| { | |
| "_name_or_path": "jiajunlong/TinyLLaVA-OpenELM-450M-SigLIP-0.89B", | |
| "architectures": [ | |
| "TinyLlavaForConditionalGeneration" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "configuration.TinyLlavaConfig", | |
| "AutoModelForCausalLM": "modeling_tinyllava_elm.TinyLlavaForConditionalGeneration" | |
| }, | |
| "cache_dir": null, | |
| "connector_type": "mlp2x_gelu", | |
| "hidden_size": 1536, | |
| "ignore_index": -100, | |
| "image_aspect_ratio": "square", | |
| "image_token_index": -200, | |
| "llm_model_name_or_path": "apple/OpenELM-450M-Instruct", | |
| "model_type": "tinyllava", | |
| "num_queries": 128, | |
| "num_resampler_layers": 3, | |
| "pad_token": "<unk>", | |
| "resampler_hidden_size": 768, | |
| "text_config": { | |
| "_name_or_path": "apple/OpenELM-450M-Instruct", | |
| "activation_fn_name": "swish", | |
| "architectures": [ | |
| "OpenELMForCausalLM" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "apple/OpenELM-450M-Instruct--configuration_openelm.OpenELMConfig", | |
| "AutoModelForCausalLM": "apple/OpenELM-450M-Instruct--modeling_openelm.OpenELMForCausalLM" | |
| }, | |
| "ffn_dim_divisor": 256, | |
| "ffn_multipliers": [ | |
| 0.5, | |
| 0.68, | |
| 0.87, | |
| 1.05, | |
| 1.24, | |
| 1.42, | |
| 1.61, | |
| 1.79, | |
| 1.97, | |
| 2.16, | |
| 2.34, | |
| 2.53, | |
| 2.71, | |
| 2.89, | |
| 3.08, | |
| 3.26, | |
| 3.45, | |
| 3.63, | |
| 3.82, | |
| 4.0 | |
| ], | |
| "ffn_with_glu": true, | |
| "head_dim": 64, | |
| "max_context_length": 2048, | |
| "model_dim": 1536, | |
| "model_type": "openelm", | |
| "normalization_layer_name": "rms_norm", | |
| "normalize_qk_projections": true, | |
| "num_gqa_groups": 4, | |
| "num_kv_heads": [ | |
| 3, | |
| 3, | |
| 3, | |
| 4, | |
| 4, | |
| 4, | |
| 4, | |
| 4, | |
| 4, | |
| 4, | |
| 5, | |
| 5, | |
| 5, | |
| 5, | |
| 5, | |
| 5, | |
| 6, | |
| 6, | |
| 6, | |
| 6 | |
| ], | |
| "num_query_heads": [ | |
| 12, | |
| 12, | |
| 12, | |
| 16, | |
| 16, | |
| 16, | |
| 16, | |
| 16, | |
| 16, | |
| 16, | |
| 20, | |
| 20, | |
| 20, | |
| 20, | |
| 20, | |
| 20, | |
| 24, | |
| 24, | |
| 24, | |
| 24 | |
| ], | |
| "num_transformer_layers": 20, | |
| "qkv_multipliers": [ | |
| 0.5, | |
| 1.0 | |
| ], | |
| "rope_freq_constant": 10000, | |
| "rope_max_length": 4096, | |
| "share_input_output_layers": true, | |
| "tie_word_embeddings": true, | |
| "torch_dtype": "float16" | |
| }, | |
| "tokenizer_model_max_length": 2048, | |
| "tokenizer_name_or_path": "meta-llama/Llama-2-7b-hf", | |
| "tokenizer_padding_side": "right", | |
| "tokenizer_use_fast": false, | |
| "torch_dtype": "float16", | |
| "transformers_version": "4.39.3", | |
| "tune_type_connector": "full", | |
| "tune_type_llm": "lora", | |
| "tune_type_vision_tower": "lora", | |
| "tune_vision_tower_from_layer": 0, | |
| "use_cache": true, | |
| "vision_config": { | |
| "hidden_act": "gelu_pytorch_tanh", | |
| "hidden_size": 1152, | |
| "image_size": 384, | |
| "intermediate_size": 4304, | |
| "layer_norm_eps": 1e-06, | |
| "model_name_or_path": "google/siglip-so400m-patch14-384", | |
| "model_name_or_path2": "", | |
| "model_type": "siglip_vision_model", | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 27, | |
| "patch_size": 14 | |
| }, | |
| "vision_feature_layer": -2, | |
| "vision_feature_select_strategy": "patch", | |
| "vision_hidden_size": 1152, | |
| "vision_model_name_or_path": "google/siglip-so400m-patch14-384", | |
| "vision_model_name_or_path2": "", | |
| "vocab_size": 32000 | |
| } | |