Image-Text-to-Text
Transformers
Safetensors
English
qwen3_5
piko
piko-9b
multimodal
vision-language
hybrid-attention
linear-attention
ocr
document-understanding
conversational
Instructions to use Dexy2/Piko-9b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Dexy2/Piko-9b with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="Dexy2/Piko-9b") 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)# Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("Dexy2/Piko-9b") model = AutoModelForMultimodalLM.from_pretrained("Dexy2/Piko-9b", 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=40) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Dexy2/Piko-9b with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Dexy2/Piko-9b" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Dexy2/Piko-9b", "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/Dexy2/Piko-9b
- SGLang
How to use Dexy2/Piko-9b 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 "Dexy2/Piko-9b" \ --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": "Dexy2/Piko-9b", "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 "Dexy2/Piko-9b" \ --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": "Dexy2/Piko-9b", "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 Dexy2/Piko-9b with Docker Model Runner:
docker model run hf.co/Dexy2/Piko-9b
| { | |
| "inputs": { | |
| "candidate": "<workspace>\\merged_model\\piko9-phase15-native-vision-base", | |
| "language": "<workspace>\\merged_model\\wraithfast-phase15-150k-full-ft", | |
| "vision": "<workspace>\\models\\qwen3.5-9b-vision", | |
| "adapters": [ | |
| "<workspace>\\adapters\\piko9-phase5-native-vision-qlora", | |
| "<workspace>\\adapters\\piko9-120k-qlora" | |
| ] | |
| }, | |
| "seed": 0, | |
| "key_sets": { | |
| "candidate_tensor_count": 775, | |
| "sources": { | |
| "language": { | |
| "tensor_count": 427, | |
| "shared_with_candidate": 427, | |
| "only_in_source": [] | |
| }, | |
| "vision": { | |
| "tensor_count": 775, | |
| "shared_with_candidate": 775, | |
| "only_in_source": [] | |
| } | |
| }, | |
| "candidate_keys_unexplained": [] | |
| }, | |
| "attribution": { | |
| "sampled": 24, | |
| "tally": { | |
| "language": 10, | |
| "language+vision": 1, | |
| "vision": 13 | |
| }, | |
| "records": [ | |
| { | |
| "key": "model.language_model.layers.1.linear_attn.in_proj_a.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 32, | |
| 4096 | |
| ] | |
| }, | |
| "matches": [ | |
| "language" | |
| ], | |
| "differs_from": [ | |
| { | |
| "source": "vision", | |
| "reason": "value", | |
| "max_abs_delta": 0.003662109375 | |
| } | |
| ], | |
| "absent_from": [], | |
| "verdict": "language" | |
| }, | |
| { | |
| "key": "model.language_model.layers.17.mlp.down_proj.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 4096, | |
| 12288 | |
| ] | |
| }, | |
| "matches": [ | |
| "language" | |
| ], | |
| "differs_from": [ | |
| { | |
| "source": "vision", | |
| "reason": "value", | |
| "max_abs_delta": 0.00341796875 | |
| } | |
| ], | |
| "absent_from": [], | |
| "verdict": "language" | |
| }, | |
| { | |
| "key": "model.language_model.layers.19.self_attn.k_norm.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 256 | |
| ] | |
| }, | |
| "matches": [ | |
| "language" | |
| ], | |
| "differs_from": [ | |
| { | |
| "source": "vision", | |
| "reason": "value", | |
| "max_abs_delta": 0.005859375 | |
| } | |
| ], | |
| "absent_from": [], | |
| "verdict": "language" | |
| }, | |
| { | |
| "key": "model.language_model.layers.21.linear_attn.norm.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 128 | |
| ] | |
| }, | |
| "matches": [ | |
| "language" | |
| ], | |
| "differs_from": [ | |
| { | |
| "source": "vision", | |
| "reason": "value", | |
| "max_abs_delta": 0.001953125 | |
| } | |
| ], | |
| "absent_from": [], | |
| "verdict": "language" | |
| }, | |
| { | |
| "key": "model.language_model.layers.22.linear_attn.in_proj_a.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 32, | |
| 4096 | |
| ] | |
| }, | |
| "matches": [ | |
| "language" | |
| ], | |
| "differs_from": [ | |
| { | |
| "source": "vision", | |
| "reason": "value", | |
| "max_abs_delta": 0.00250244140625 | |
| } | |
| ], | |
| "absent_from": [], | |
| "verdict": "language" | |
| }, | |
| { | |
| "key": "model.language_model.layers.22.mlp.up_proj.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 12288, | |
| 4096 | |
| ] | |
| }, | |
| "matches": [ | |
| "language" | |
| ], | |
| "differs_from": [ | |
| { | |
| "source": "vision", | |
| "reason": "value", | |
| "max_abs_delta": 0.006195068359375 | |
| } | |
| ], | |
| "absent_from": [], | |
| "verdict": "language" | |
| }, | |
| { | |
| "key": "model.language_model.layers.25.linear_attn.in_proj_qkv.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 8192, | |
| 4096 | |
| ] | |
| }, | |
| "matches": [ | |
| "language" | |
| ], | |
| "differs_from": [ | |
| { | |
| "source": "vision", | |
| "reason": "value", | |
| "max_abs_delta": 0.00286865234375 | |
| } | |
| ], | |
| "absent_from": [], | |
| "verdict": "language" | |
| }, | |
| { | |
| "key": "model.language_model.layers.26.linear_attn.in_proj_b.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 32, | |
| 4096 | |
| ] | |
| }, | |
| "matches": [ | |
| "language" | |
| ], | |
| "differs_from": [ | |
| { | |
| "source": "vision", | |
| "reason": "value", | |
| "max_abs_delta": 0.002288818359375 | |
| } | |
| ], | |
| "absent_from": [], | |
| "verdict": "language" | |
| }, | |
| { | |
| "key": "model.language_model.layers.7.mlp.down_proj.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 4096, | |
| 12288 | |
| ] | |
| }, | |
| "matches": [ | |
| "language" | |
| ], | |
| "differs_from": [ | |
| { | |
| "source": "vision", | |
| "reason": "value", | |
| "max_abs_delta": 0.002685546875 | |
| } | |
| ], | |
| "absent_from": [], | |
| "verdict": "language" | |
| }, | |
| { | |
| "key": "model.language_model.layers.8.linear_attn.dt_bias", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 32 | |
| ] | |
| }, | |
| "matches": [ | |
| "language", | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [], | |
| "verdict": "language+vision" | |
| }, | |
| { | |
| "key": "model.language_model.layers.9.mlp.up_proj.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 12288, | |
| 4096 | |
| ] | |
| }, | |
| "matches": [ | |
| "language" | |
| ], | |
| "differs_from": [ | |
| { | |
| "source": "vision", | |
| "reason": "value", | |
| "max_abs_delta": 0.0050048828125 | |
| } | |
| ], | |
| "absent_from": [], | |
| "verdict": "language" | |
| }, | |
| { | |
| "key": "model.visual.blocks.12.attn.proj.bias", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 1152 | |
| ] | |
| }, | |
| "matches": [ | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [ | |
| "language" | |
| ], | |
| "verdict": "vision" | |
| }, | |
| { | |
| "key": "model.visual.blocks.13.norm2.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 1152 | |
| ] | |
| }, | |
| "matches": [ | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [ | |
| "language" | |
| ], | |
| "verdict": "vision" | |
| }, | |
| { | |
| "key": "model.visual.blocks.17.attn.qkv.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 3456, | |
| 1152 | |
| ] | |
| }, | |
| "matches": [ | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [ | |
| "language" | |
| ], | |
| "verdict": "vision" | |
| }, | |
| { | |
| "key": "model.visual.blocks.18.norm1.bias", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 1152 | |
| ] | |
| }, | |
| "matches": [ | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [ | |
| "language" | |
| ], | |
| "verdict": "vision" | |
| }, | |
| { | |
| "key": "model.visual.blocks.2.attn.proj.bias", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 1152 | |
| ] | |
| }, | |
| "matches": [ | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [ | |
| "language" | |
| ], | |
| "verdict": "vision" | |
| }, | |
| { | |
| "key": "model.visual.blocks.22.attn.qkv.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 3456, | |
| 1152 | |
| ] | |
| }, | |
| "matches": [ | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [ | |
| "language" | |
| ], | |
| "verdict": "vision" | |
| }, | |
| { | |
| "key": "model.visual.blocks.3.mlp.linear_fc1.bias", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 4304 | |
| ] | |
| }, | |
| "matches": [ | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [ | |
| "language" | |
| ], | |
| "verdict": "vision" | |
| }, | |
| { | |
| "key": "model.visual.blocks.4.mlp.linear_fc2.bias", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 1152 | |
| ] | |
| }, | |
| "matches": [ | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [ | |
| "language" | |
| ], | |
| "verdict": "vision" | |
| }, | |
| { | |
| "key": "model.visual.blocks.5.norm1.bias", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 1152 | |
| ] | |
| }, | |
| "matches": [ | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [ | |
| "language" | |
| ], | |
| "verdict": "vision" | |
| }, | |
| { | |
| "key": "model.visual.blocks.7.norm2.bias", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 1152 | |
| ] | |
| }, | |
| "matches": [ | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [ | |
| "language" | |
| ], | |
| "verdict": "vision" | |
| }, | |
| { | |
| "key": "model.visual.blocks.8.norm1.bias", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 1152 | |
| ] | |
| }, | |
| "matches": [ | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [ | |
| "language" | |
| ], | |
| "verdict": "vision" | |
| }, | |
| { | |
| "key": "model.visual.blocks.9.mlp.linear_fc1.bias", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 4304 | |
| ] | |
| }, | |
| "matches": [ | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [ | |
| "language" | |
| ], | |
| "verdict": "vision" | |
| }, | |
| { | |
| "key": "mtp.layers.0.post_attention_layernorm.weight", | |
| "meta": { | |
| "dtype": "BF16", | |
| "shape": [ | |
| 4096 | |
| ] | |
| }, | |
| "matches": [ | |
| "vision" | |
| ], | |
| "differs_from": [], | |
| "absent_from": [ | |
| "language" | |
| ], | |
| "verdict": "vision" | |
| } | |
| ] | |
| }, | |
| "adapters": [ | |
| { | |
| "adapter": "piko9-phase5-native-vision-qlora", | |
| "status": "checked", | |
| "rank": 32, | |
| "lora_alpha": 64, | |
| "scaling": 2.0, | |
| "base_model_name_or_path": "merged_model/piko9-phase15-native-vision-base", | |
| "lora_module_count": 128, | |
| "modules_checked": 6, | |
| "votes": { | |
| "merged": 0, | |
| "not_merged": 6, | |
| "inconclusive": 0 | |
| }, | |
| "conclusion": "adapter_is_NOT_present_in_published_weights", | |
| "checks": [ | |
| { | |
| "module": "base_model.model.model.language_model.layers.0.mlp.down_proj", | |
| "candidate_tensor": "model.language_model.layers.0.mlp.down_proj.weight", | |
| "lora_delta_absmax": 0.0026623515877872705, | |
| "absmax_candidate_minus_language": 0.0, | |
| "absmax_candidate_minus_language_plus_delta": 0.002662351354956627, | |
| "verdict": "not_merged" | |
| }, | |
| { | |
| "module": "base_model.model.model.language_model.layers.0.mlp.gate_proj", | |
| "candidate_tensor": "model.language_model.layers.0.mlp.gate_proj.weight", | |
| "lora_delta_absmax": 0.0067080664448440075, | |
| "absmax_candidate_minus_language": 0.0, | |
| "absmax_candidate_minus_language_plus_delta": 0.006708066910505295, | |
| "verdict": "not_merged" | |
| }, | |
| { | |
| "module": "base_model.model.model.language_model.layers.0.mlp.up_proj", | |
| "candidate_tensor": "model.language_model.layers.0.mlp.up_proj.weight", | |
| "lora_delta_absmax": 0.004918927326798439, | |
| "absmax_candidate_minus_language": 0.0, | |
| "absmax_candidate_minus_language_plus_delta": 0.004918927326798439, | |
| "verdict": "not_merged" | |
| }, | |
| { | |
| "module": "base_model.model.model.language_model.layers.1.mlp.down_proj", | |
| "candidate_tensor": "model.language_model.layers.1.mlp.down_proj.weight", | |
| "lora_delta_absmax": 0.0016279240371659398, | |
| "absmax_candidate_minus_language": 0.0, | |
| "absmax_candidate_minus_language_plus_delta": 0.001627923920750618, | |
| "verdict": "not_merged" | |
| }, | |
| { | |
| "module": "base_model.model.model.language_model.layers.1.mlp.gate_proj", | |
| "candidate_tensor": "model.language_model.layers.1.mlp.gate_proj.weight", | |
| "lora_delta_absmax": 0.005791253875941038, | |
| "absmax_candidate_minus_language": 0.0, | |
| "absmax_candidate_minus_language_plus_delta": 0.005791253875941038, | |
| "verdict": "not_merged" | |
| }, | |
| { | |
| "module": "base_model.model.model.language_model.layers.1.mlp.up_proj", | |
| "candidate_tensor": "model.language_model.layers.1.mlp.up_proj.weight", | |
| "lora_delta_absmax": 0.005112477578222752, | |
| "absmax_candidate_minus_language": 0.0, | |
| "absmax_candidate_minus_language_plus_delta": 0.005112477578222752, | |
| "verdict": "not_merged" | |
| } | |
| ] | |
| }, | |
| { | |
| "adapter": "piko9-120k-qlora", | |
| "status": "checked", | |
| "rank": 32, | |
| "lora_alpha": 64, | |
| "scaling": 2.0, | |
| "base_model_name_or_path": "merged_model/wraithfast-phase15-150k-full-ft", | |
| "lora_module_count": 128, | |
| "modules_checked": 6, | |
| "votes": { | |
| "merged": 0, | |
| "not_merged": 6, | |
| "inconclusive": 0 | |
| }, | |
| "conclusion": "adapter_is_NOT_present_in_published_weights", | |
| "checks": [ | |
| { | |
| "module": "base_model.model.model.layers.0.mlp.down_proj", | |
| "candidate_tensor": "model.language_model.layers.0.mlp.down_proj.weight", | |
| "lora_delta_absmax": 0.0026120361872017384, | |
| "absmax_candidate_minus_language": 0.0, | |
| "absmax_candidate_minus_language_plus_delta": 0.002612035721540451, | |
| "verdict": "not_merged" | |
| }, | |
| { | |
| "module": "base_model.model.model.layers.0.mlp.gate_proj", | |
| "candidate_tensor": "model.language_model.layers.0.mlp.gate_proj.weight", | |
| "lora_delta_absmax": 0.005998868029564619, | |
| "absmax_candidate_minus_language": 0.0, | |
| "absmax_candidate_minus_language_plus_delta": 0.005998868495225906, | |
| "verdict": "not_merged" | |
| }, | |
| { | |
| "module": "base_model.model.model.layers.0.mlp.up_proj", | |
| "candidate_tensor": "model.language_model.layers.0.mlp.up_proj.weight", | |
| "lora_delta_absmax": 0.004918788559734821, | |
| "absmax_candidate_minus_language": 0.0, | |
| "absmax_candidate_minus_language_plus_delta": 0.004918788559734821, | |
| "verdict": "not_merged" | |
| }, | |
| { | |
| "module": "base_model.model.model.layers.1.mlp.down_proj", | |
| "candidate_tensor": "model.language_model.layers.1.mlp.down_proj.weight", | |
| "lora_delta_absmax": 0.0015581832267343998, | |
| "absmax_candidate_minus_language": 0.0, | |
| "absmax_candidate_minus_language_plus_delta": 0.0015581832267343998, | |
| "verdict": "not_merged" | |
| }, | |
| { | |
| "module": "base_model.model.model.layers.1.mlp.gate_proj", | |
| "candidate_tensor": "model.language_model.layers.1.mlp.gate_proj.weight", | |
| "lora_delta_absmax": 0.005982093047350645, | |
| "absmax_candidate_minus_language": 0.0, | |
| "absmax_candidate_minus_language_plus_delta": 0.005982093047350645, | |
| "verdict": "not_merged" | |
| }, | |
| { | |
| "module": "base_model.model.model.layers.1.mlp.up_proj", | |
| "candidate_tensor": "model.language_model.layers.1.mlp.up_proj.weight", | |
| "lora_delta_absmax": 0.004832717124372721, | |
| "absmax_candidate_minus_language": 0.0, | |
| "absmax_candidate_minus_language_plus_delta": 0.004832716658711433, | |
| "verdict": "not_merged" | |
| } | |
| ] | |
| } | |
| ] | |
| } | |