Image-Text-to-Text
Transformers
Safetensors
English
qwen3_5
gptq
int4
4-bit precision
qwen3.5
quantization
agentic-coding
reasoning
conversational
Instructions to use malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4") 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("malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4") model = AutoModelForMultimodalLM.from_pretrained("malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4", 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 malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4", "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/malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4
- SGLang
How to use malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4 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 "malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4" \ --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": "malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4", "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 "malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4" \ --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": "malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4", "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 malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4 with Docker Model Runner:
docker model run hf.co/malvavisc0/Qwable-9B-Claude-Fable-5-GPTQ-Int4
| layer,module,loss,samples,damp,time | |
| 0,linear_attn.in_proj_qkv,0.0002923562,0.05000,0.757 | |
| 0,linear_attn.in_proj_z,0.0001466612,0.05000,0.443 | |
| 0,linear_attn.out_proj,0.0000000849,0.05000,0.441 | |
| 0,mlp.up_proj,0.0000033608,0.05000,1.322 | |
| 0,mlp.gate_proj,0.0000038972,0.05000,1.313 | |
| 0,mlp.down_proj,0.0000000298,0.05000,2.017 | |
| 1,linear_attn.in_proj_qkv,0.0000253694,0.05000,0.507 | |
| 1,linear_attn.in_proj_z,0.0000127770,0.05000,0.446 | |
| 1,linear_attn.out_proj,0.0000000393,0.05000,0.447 | |
| 1,mlp.up_proj,0.0000081792,0.05000,1.538 | |
| 1,mlp.gate_proj,0.0000091871,0.05000,1.567 | |
| 1,mlp.down_proj,0.0000000763,0.05000,1.959 | |
| 2,linear_attn.in_proj_qkv,0.0000438525,0.05000,0.508 | |
| 2,linear_attn.in_proj_z,0.0000216663,0.05000,0.444 | |
| 2,linear_attn.out_proj,0.0000000682,0.05000,0.444 | |
| 2,mlp.gate_proj,0.0000150929,0.05000,1.324 | |
| 2,mlp.up_proj,0.0000126798,0.05000,1.331 | |
| 2,mlp.down_proj,0.0000001957,0.05000,1.974 | |
| 3,self_attn.k_proj,0.0000192607,0.05000,1.806 | |
| 3,self_attn.q_proj,0.0001859263,0.05000,1.821 | |
| 3,self_attn.v_proj,0.0000181077,0.05000,1.834 | |
| 3,self_attn.o_proj,0.0000001161,0.05000,0.441 | |
| 3,mlp.up_proj,0.0000181847,0.05000,1.498 | |
| 3,mlp.gate_proj,0.0000201043,0.05000,1.502 | |
| 3,mlp.down_proj,0.0000002504,0.05000,1.937 | |
| 4,linear_attn.in_proj_qkv,0.0000795212,0.05000,0.509 | |
| 4,linear_attn.in_proj_z,0.0000402917,0.05000,0.448 | |
| 4,linear_attn.out_proj,0.0000001660,0.05000,0.440 | |
| 4,mlp.up_proj,0.0000272368,0.05000,1.245 | |
| 4,mlp.gate_proj,0.0000302054,0.05000,1.251 | |
| 4,mlp.down_proj,0.0000004177,0.05000,1.985 | |
| 5,linear_attn.in_proj_qkv,0.0001058409,0.05000,0.507 | |
| 5,linear_attn.in_proj_z,0.0000515063,0.05000,0.443 | |
| 5,linear_attn.out_proj,0.0000003372,0.05000,0.441 | |
| 5,mlp.up_proj,0.0000352061,0.05000,1.595 | |
| 5,mlp.gate_proj,0.0000418450,0.05000,1.607 | |
| 5,mlp.down_proj,0.0000007811,0.05000,1.945 | |
| 6,linear_attn.in_proj_qkv,0.0001155280,0.05000,0.504 | |
| 6,linear_attn.in_proj_z,0.0000475952,0.05000,0.445 | |
| 6,linear_attn.out_proj,0.0000007005,0.05000,0.444 | |
| 6,mlp.gate_proj,0.0000575008,0.05000,1.463 | |
| 6,mlp.up_proj,0.0000465150,0.05000,1.484 | |
| 6,mlp.down_proj,0.0000013511,0.05000,1.962 | |
| 7,self_attn.k_proj,0.0000173141,0.05000,1.807 | |
| 7,self_attn.q_proj,0.0001302608,0.05000,1.814 | |
| 7,self_attn.v_proj,0.0000149477,0.05000,1.827 | |
| 7,self_attn.o_proj,0.0000008519,0.05000,0.450 | |
| 7,mlp.up_proj,0.0000492923,0.05000,1.635 | |
| 7,mlp.gate_proj,0.0000618966,0.05000,1.636 | |
| 7,mlp.down_proj,0.0000012843,0.05000,1.960 | |
| 8,linear_attn.in_proj_qkv,0.0001362903,0.05000,0.506 | |
| 8,linear_attn.in_proj_z,0.0000582160,0.05000,0.443 | |
| 8,linear_attn.out_proj,0.0000007473,0.05000,0.448 | |
| 8,mlp.gate_proj,0.0000654837,0.05000,1.272 | |
| 8,mlp.up_proj,0.0000539881,0.05000,1.283 | |
| 8,mlp.down_proj,0.0000013767,0.05000,1.944 | |
| 9,linear_attn.in_proj_qkv,0.0001456481,0.05000,0.509 | |
| 9,linear_attn.in_proj_z,0.0000594048,0.05000,0.443 | |
| 9,linear_attn.out_proj,0.0000008285,0.05000,0.446 | |
| 9,mlp.gate_proj,0.0000664895,0.05000,1.591 | |
| 9,mlp.up_proj,0.0000585528,0.05000,1.611 | |
| 9,mlp.down_proj,0.0000015177,0.05000,1.943 | |
| 10,linear_attn.in_proj_qkv,0.0001391888,0.05000,0.505 | |
| 10,linear_attn.in_proj_z,0.0000545197,0.05000,0.440 | |
| 10,linear_attn.out_proj,0.0000009204,0.05000,0.448 | |
| 10,mlp.up_proj,0.0000611052,0.05000,1.330 | |
| 10,mlp.gate_proj,0.0000634157,0.05000,1.347 | |
| 10,mlp.down_proj,0.0000016957,0.05000,1.938 | |
| 11,self_attn.k_proj,0.0000148096,0.05000,1.792 | |
| 11,self_attn.q_proj,0.0001025371,0.05000,1.802 | |
| 11,self_attn.v_proj,0.0000139872,0.05000,1.817 | |
| 11,self_attn.o_proj,0.0000014906,0.05000,0.443 | |
| 11,mlp.up_proj,0.0000636308,0.05000,1.240 | |
| 11,mlp.gate_proj,0.0000622587,0.05000,1.257 | |
| 11,mlp.down_proj,0.0000017404,0.05000,1.934 | |
| 12,linear_attn.in_proj_qkv,0.0001382742,0.05000,0.506 | |
| 12,linear_attn.in_proj_z,0.0000501939,0.05000,0.450 | |
| 12,linear_attn.out_proj,0.0000011899,0.05000,0.448 | |
| 12,mlp.gate_proj,0.0000613396,0.05000,1.444 | |
| 12,mlp.up_proj,0.0000652184,0.05000,1.475 | |
| 12,mlp.down_proj,0.0000018146,0.05000,1.946 | |
| 13,linear_attn.in_proj_qkv,0.0001527687,0.05000,0.509 | |
| 13,linear_attn.in_proj_z,0.0000576133,0.05000,0.435 | |
| 13,linear_attn.out_proj,0.0000011374,0.05000,0.447 | |
| 13,mlp.up_proj,0.0000669694,0.05000,1.612 | |
| 13,mlp.gate_proj,0.0000618521,0.05000,1.628 | |
| 13,mlp.down_proj,0.0000019409,0.05000,1.931 | |
| 14,linear_attn.in_proj_qkv,0.0001479856,0.05000,0.511 | |
| 14,linear_attn.in_proj_z,0.0000557005,0.05000,0.443 | |
| 14,linear_attn.out_proj,0.0000018432,0.05000,0.447 | |
| 14,mlp.gate_proj,0.0000602646,0.05000,1.386 | |
| 14,mlp.up_proj,0.0000687781,0.05000,1.395 | |
| 14,mlp.down_proj,0.0000021436,0.05000,1.947 | |
| 15,self_attn.q_proj,0.0001035363,0.05000,1.816 | |
| 15,self_attn.v_proj,0.0000182662,0.05000,1.835 | |
| 15,self_attn.k_proj,0.0000149197,0.05000,1.844 | |
| 15,self_attn.o_proj,0.0000017794,0.05000,0.440 | |
| 15,mlp.gate_proj,0.0000654744,0.05000,1.591 | |
| 15,mlp.up_proj,0.0000758326,0.05000,1.606 | |
| 15,mlp.down_proj,0.0000024666,0.05000,1.959 | |
| 16,linear_attn.in_proj_qkv,0.0001525745,0.05000,0.506 | |
| 16,linear_attn.in_proj_z,0.0000568288,0.05000,0.449 | |
| 16,linear_attn.out_proj,0.0000015897,0.05000,0.442 | |
| 16,mlp.up_proj,0.0000812780,0.05000,1.391 | |
| 16,mlp.gate_proj,0.0000707985,0.05000,1.402 | |
| 16,mlp.down_proj,0.0000029527,0.05000,1.959 | |
| 17,linear_attn.in_proj_qkv,0.0002041090,0.05000,0.503 | |
| 17,linear_attn.in_proj_z,0.0000627605,0.05000,0.442 | |
| 17,linear_attn.out_proj,0.0000018685,0.05000,0.439 | |
| 17,mlp.up_proj,0.0000876574,0.05000,1.601 | |
| 17,mlp.gate_proj,0.0000753859,0.05000,1.619 | |
| 17,mlp.down_proj,0.0000039025,0.05000,1.941 | |
| 18,linear_attn.in_proj_qkv,0.0001874450,0.05000,0.501 | |
| 18,linear_attn.in_proj_z,0.0000663500,0.05000,0.450 | |
| 18,linear_attn.out_proj,0.0000032130,0.05000,0.438 | |
| 18,mlp.gate_proj,0.0000935092,0.05000,1.572 | |
| 18,mlp.up_proj,0.0001019622,0.05000,1.592 | |
| 18,mlp.down_proj,0.0000082581,0.05000,1.961 | |
| 19,self_attn.k_proj,0.0000220895,0.05000,1.790 | |
| 19,self_attn.q_proj,0.0001441360,0.05000,1.805 | |
| 19,self_attn.v_proj,0.0000465971,0.05000,1.823 | |
| 19,self_attn.o_proj,0.0000064091,0.05000,0.450 | |
| 19,mlp.gate_proj,0.0001120633,0.05000,1.334 | |
| 19,mlp.up_proj,0.0001273858,0.05000,1.355 | |
| 19,mlp.down_proj,0.0000112897,0.05000,1.961 | |
| 20,linear_attn.in_proj_qkv,0.0003139259,0.05000,0.506 | |
| 20,linear_attn.in_proj_z,0.0001078105,0.05000,0.441 | |
| 20,linear_attn.out_proj,0.0000030580,0.05000,0.445 | |
| 20,mlp.up_proj,0.0001373163,0.05000,1.439 | |
| 20,mlp.gate_proj,0.0001436044,0.05000,1.448 | |
| 20,mlp.down_proj,0.0000117298,0.05000,1.940 | |
| 21,linear_attn.in_proj_qkv,0.0003123837,0.05000,0.506 | |
| 21,linear_attn.in_proj_z,0.0001131420,0.05000,0.447 | |
| 21,linear_attn.out_proj,0.0000044404,0.05000,0.439 | |
| 21,mlp.up_proj,0.0001441978,0.05000,1.432 | |
| 21,mlp.gate_proj,0.0001608396,0.05000,1.442 | |
| 21,mlp.down_proj,0.0000144337,0.05000,1.969 | |
| 22,linear_attn.in_proj_qkv,0.0003056657,0.05000,0.500 | |
| 22,linear_attn.in_proj_z,0.0001269550,0.05000,0.450 | |
| 22,linear_attn.out_proj,0.0000055819,0.05000,0.448 | |
| 22,mlp.gate_proj,0.0002304516,0.05000,1.337 | |
| 22,mlp.up_proj,0.0001867799,0.05000,1.342 | |
| 22,mlp.down_proj,0.0000269420,0.05000,1.964 | |
| 23,self_attn.v_proj,0.0000701680,0.05000,1.793 | |
| 23,self_attn.q_proj,0.0002147914,0.05000,1.801 | |
| 23,self_attn.k_proj,0.0000358240,0.05000,1.807 | |
| 23,self_attn.o_proj,0.0000100568,0.05000,0.445 | |
| 23,mlp.up_proj,0.0001871019,0.05000,1.242 | |
| 23,mlp.gate_proj,0.0002105997,0.05000,1.246 | |
| 23,mlp.down_proj,0.0000214708,0.05000,1.961 | |
| 24,linear_attn.in_proj_qkv,0.0003712067,0.05000,0.504 | |
| 24,linear_attn.in_proj_z,0.0001371180,0.05000,0.440 | |
| 24,linear_attn.out_proj,0.0000084203,0.05000,0.441 | |
| 24,mlp.up_proj,0.0001981273,0.05000,1.262 | |
| 24,mlp.gate_proj,0.0002350082,0.05000,1.264 | |
| 24,mlp.down_proj,0.0000208720,0.05000,1.947 | |
| 25,linear_attn.in_proj_qkv,0.0004077492,0.05000,0.507 | |
| 25,linear_attn.in_proj_z,0.0001532642,0.05000,0.447 | |
| 25,linear_attn.out_proj,0.0000076455,0.05000,0.445 | |
| 25,mlp.gate_proj,0.0002800821,0.05000,1.241 | |
| 25,mlp.up_proj,0.0002269588,0.05000,1.247 | |
| 25,mlp.down_proj,0.0000223139,0.05000,1.959 | |
| 26,linear_attn.in_proj_qkv,0.0003774328,0.05000,0.509 | |
| 26,linear_attn.in_proj_z,0.0001634729,0.05000,0.437 | |
| 26,linear_attn.out_proj,0.0000086689,0.05000,0.440 | |
| 26,mlp.up_proj,0.0002674957,0.05000,1.266 | |
| 26,mlp.gate_proj,0.0003367386,0.05000,1.272 | |
| 26,mlp.down_proj,0.0000294222,0.05000,1.934 | |
| 27,self_attn.k_proj,0.0000531112,0.05000,1.798 | |
| 27,self_attn.v_proj,0.0001838021,0.05000,1.835 | |
| 27,self_attn.q_proj,0.0002792344,0.05000,1.834 | |
| 27,self_attn.o_proj,0.0000320682,0.05000,0.438 | |
| 27,mlp.up_proj,0.0002780721,0.05000,1.264 | |
| 27,mlp.gate_proj,0.0003418540,0.05000,1.281 | |
| 27,mlp.down_proj,0.0000346843,0.05000,1.955 | |
| 28,linear_attn.in_proj_qkv,0.0004718472,0.05000,0.497 | |
| 28,linear_attn.in_proj_z,0.0001779310,0.05000,0.449 | |
| 28,linear_attn.out_proj,0.0000171880,0.05000,0.442 | |
| 28,mlp.gate_proj,0.0003511496,0.05000,1.426 | |
| 28,mlp.up_proj,0.0002891008,0.05000,1.435 | |
| 28,mlp.down_proj,0.0000412876,0.05000,1.966 | |
| 29,linear_attn.in_proj_qkv,0.0003411556,0.05000,0.505 | |
| 29,linear_attn.in_proj_z,0.0001564033,0.05000,0.439 | |
| 29,linear_attn.out_proj,0.0000169771,0.05000,0.447 | |
| 29,mlp.gate_proj,0.0003809890,0.05000,1.373 | |
| 29,mlp.up_proj,0.0003110860,0.05000,1.395 | |
| 29,mlp.down_proj,0.0000492641,0.05000,1.937 | |
| 30,linear_attn.in_proj_qkv,0.0003637233,0.05000,0.511 | |
| 30,linear_attn.in_proj_z,0.0001553133,0.05000,0.451 | |
| 30,linear_attn.out_proj,0.0000314266,0.05000,0.451 | |
| 30,mlp.up_proj,0.0002988835,0.05000,1.609 | |
| 30,mlp.gate_proj,0.0003591503,0.05000,1.618 | |
| 30,mlp.down_proj,0.0000692388,0.05000,1.969 | |
| 31,self_attn.v_proj,0.0001614979,0.05000,1.466 | |
| 31,self_attn.k_proj,0.0000528025,0.05000,1.486 | |
| 31,self_attn.q_proj,0.0002747820,0.05000,1.487 | |
| 31,self_attn.o_proj,0.0000677047,0.05000,0.453 | |
| 31,mlp.gate_proj,0.0003321292,0.05000,1.419 | |
| 31,mlp.up_proj,0.0002685991,0.05000,1.428 | |
| 31,mlp.down_proj,0.0001504178,0.05000,1.965 | |