Image-Text-to-Text
Transformers
Safetensors
paligemma
text-generation-inference
8-bit precision
bitsandbytes
Instructions to use Pushpendra817/Xtraektor with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Pushpendra817/Xtraektor with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="Pushpendra817/Xtraektor")# Load model directly from transformers import AutoProcessor, AutoModelForImageTextToText processor = AutoProcessor.from_pretrained("Pushpendra817/Xtraektor") model = AutoModelForImageTextToText.from_pretrained("Pushpendra817/Xtraektor") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Pushpendra817/Xtraektor with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Pushpendra817/Xtraektor" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Pushpendra817/Xtraektor", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Pushpendra817/Xtraektor
- SGLang
How to use Pushpendra817/Xtraektor 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 "Pushpendra817/Xtraektor" \ --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": "Pushpendra817/Xtraektor", "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 "Pushpendra817/Xtraektor" \ --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": "Pushpendra817/Xtraektor", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Pushpendra817/Xtraektor with Docker Model Runner:
docker model run hf.co/Pushpendra817/Xtraektor
Upload config.json
Browse files- config.json +55 -0
config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/paligemma-3b-mix-224",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"PaliGemmaForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"bos_token_id": 2,
|
| 7 |
+
"eos_token_id": 1,
|
| 8 |
+
"hidden_size": 2048,
|
| 9 |
+
"ignore_index": -100,
|
| 10 |
+
"image_token_index": 257152,
|
| 11 |
+
"model_type": "paligemma",
|
| 12 |
+
"pad_token_id": 0,
|
| 13 |
+
"projection_dim": 2048,
|
| 14 |
+
"quantization_config": {
|
| 15 |
+
"_load_in_4bit": false,
|
| 16 |
+
"_load_in_8bit": true,
|
| 17 |
+
"bnb_4bit_compute_dtype": "float32",
|
| 18 |
+
"bnb_4bit_quant_storage": "uint8",
|
| 19 |
+
"bnb_4bit_quant_type": "fp4",
|
| 20 |
+
"bnb_4bit_use_double_quant": false,
|
| 21 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
| 22 |
+
"llm_int8_has_fp16_weight": false,
|
| 23 |
+
"llm_int8_skip_modules": null,
|
| 24 |
+
"llm_int8_threshold": 6.0,
|
| 25 |
+
"load_in_4bit": false,
|
| 26 |
+
"load_in_8bit": true,
|
| 27 |
+
"quant_method": "bitsandbytes"
|
| 28 |
+
},
|
| 29 |
+
"text_config": {
|
| 30 |
+
"hidden_size": 2048,
|
| 31 |
+
"intermediate_size": 16384,
|
| 32 |
+
"model_type": "gemma",
|
| 33 |
+
"num_attention_heads": 8,
|
| 34 |
+
"num_hidden_layers": 18,
|
| 35 |
+
"num_image_tokens": 256,
|
| 36 |
+
"num_key_value_heads": 1,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"vocab_size": 257216
|
| 39 |
+
},
|
| 40 |
+
"torch_dtype": "float16",
|
| 41 |
+
"transformers_version": "4.42.4",
|
| 42 |
+
"vision_config": {
|
| 43 |
+
"hidden_size": 1152,
|
| 44 |
+
"intermediate_size": 4304,
|
| 45 |
+
"model_type": "siglip_vision_model",
|
| 46 |
+
"num_attention_heads": 16,
|
| 47 |
+
"num_hidden_layers": 27,
|
| 48 |
+
"num_image_tokens": 256,
|
| 49 |
+
"patch_size": 14,
|
| 50 |
+
"projection_dim": 2048,
|
| 51 |
+
"projector_hidden_act": "gelu_fast",
|
| 52 |
+
"vision_use_head": false
|
| 53 |
+
},
|
| 54 |
+
"vocab_size": 257216
|
| 55 |
+
}
|