Instructions to use trohrbaugh/Qwen3.5-35B-A3B-heretic with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use trohrbaugh/Qwen3.5-35B-A3B-heretic with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="trohrbaugh/Qwen3.5-35B-A3B-heretic")
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("trohrbaugh/Qwen3.5-35B-A3B-heretic")
model = AutoModelForMultimodalLM.from_pretrained("trohrbaugh/Qwen3.5-35B-A3B-heretic", 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 trohrbaugh/Qwen3.5-35B-A3B-heretic with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "trohrbaugh/Qwen3.5-35B-A3B-heretic"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "trohrbaugh/Qwen3.5-35B-A3B-heretic",
		"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/trohrbaugh/Qwen3.5-35B-A3B-heretic

SGLang

How to use trohrbaugh/Qwen3.5-35B-A3B-heretic 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 "trohrbaugh/Qwen3.5-35B-A3B-heretic" \
    --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": "trohrbaugh/Qwen3.5-35B-A3B-heretic",
		"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 "trohrbaugh/Qwen3.5-35B-A3B-heretic" \
        --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": "trohrbaugh/Qwen3.5-35B-A3B-heretic",
		"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 trohrbaugh/Qwen3.5-35B-A3B-heretic with Docker Model Runner:
```
docker model run hf.co/trohrbaugh/Qwen3.5-35B-A3B-heretic
```

trohrbaugh commited on May 2

Commit

cec496e

verified ·

1 Parent(s): 316c8e6

moe-fuse: register 80 fused keys

Browse files

Files changed (1) hide show

model.safetensors.index.json +81 -1

model.safetensors.index.json CHANGED Viewed

@@ -32454,6 +32454,86 @@
     "mtp.layers.0.self_attn.v_proj.weight": "model-auxiliary.safetensors",
     "mtp.norm.weight": "model-auxiliary.safetensors",
     "mtp.pre_fc_norm_embedding.weight": "model-auxiliary.safetensors",
-    "mtp.pre_fc_norm_hidden.weight": "model-auxiliary.safetensors"
   }
 }

     "mtp.layers.0.self_attn.v_proj.weight": "model-auxiliary.safetensors",
     "mtp.norm.weight": "model-auxiliary.safetensors",
     "mtp.pre_fc_norm_embedding.weight": "model-auxiliary.safetensors",
+    "mtp.pre_fc_norm_hidden.weight": "model-auxiliary.safetensors",
+    "model.language_model.layers.0.mlp.experts.gate_up_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.0.mlp.experts.down_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.1.mlp.experts.gate_up_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.1.mlp.experts.down_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.10.mlp.experts.gate_up_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.10.mlp.experts.down_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.11.mlp.experts.gate_up_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.11.mlp.experts.down_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.12.mlp.experts.gate_up_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.12.mlp.experts.down_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.13.mlp.experts.gate_up_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.13.mlp.experts.down_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.14.mlp.experts.gate_up_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.14.mlp.experts.down_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.15.mlp.experts.gate_up_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.15.mlp.experts.down_proj": "model-moe-fused-000.safetensors",
+    "model.language_model.layers.16.mlp.experts.gate_up_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.16.mlp.experts.down_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.17.mlp.experts.gate_up_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.17.mlp.experts.down_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.18.mlp.experts.gate_up_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.18.mlp.experts.down_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.19.mlp.experts.gate_up_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.19.mlp.experts.down_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.2.mlp.experts.gate_up_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.2.mlp.experts.down_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.20.mlp.experts.gate_up_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.20.mlp.experts.down_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.21.mlp.experts.gate_up_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.21.mlp.experts.down_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.22.mlp.experts.gate_up_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.22.mlp.experts.down_proj": "model-moe-fused-001.safetensors",
+    "model.language_model.layers.23.mlp.experts.gate_up_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.23.mlp.experts.down_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.24.mlp.experts.gate_up_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.24.mlp.experts.down_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.25.mlp.experts.gate_up_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.25.mlp.experts.down_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.26.mlp.experts.gate_up_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.26.mlp.experts.down_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.27.mlp.experts.gate_up_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.27.mlp.experts.down_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.28.mlp.experts.gate_up_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.28.mlp.experts.down_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.29.mlp.experts.gate_up_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.29.mlp.experts.down_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.3.mlp.experts.gate_up_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.3.mlp.experts.down_proj": "model-moe-fused-002.safetensors",
+    "model.language_model.layers.30.mlp.experts.gate_up_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.30.mlp.experts.down_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.31.mlp.experts.gate_up_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.31.mlp.experts.down_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.32.mlp.experts.gate_up_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.32.mlp.experts.down_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.33.mlp.experts.gate_up_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.33.mlp.experts.down_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.34.mlp.experts.gate_up_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.34.mlp.experts.down_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.35.mlp.experts.gate_up_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.35.mlp.experts.down_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.36.mlp.experts.gate_up_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.36.mlp.experts.down_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.37.mlp.experts.gate_up_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.37.mlp.experts.down_proj": "model-moe-fused-003.safetensors",
+    "model.language_model.layers.38.mlp.experts.gate_up_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.38.mlp.experts.down_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.39.mlp.experts.gate_up_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.39.mlp.experts.down_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.4.mlp.experts.gate_up_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.4.mlp.experts.down_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.5.mlp.experts.gate_up_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.5.mlp.experts.down_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.6.mlp.experts.gate_up_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.6.mlp.experts.down_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.7.mlp.experts.gate_up_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.7.mlp.experts.down_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.8.mlp.experts.gate_up_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.8.mlp.experts.down_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.9.mlp.experts.gate_up_proj": "model-moe-fused-004.safetensors",
+    "model.language_model.layers.9.mlp.experts.down_proj": "model-moe-fused-004.safetensors"
   }
 }