Instructions to use SmallDoge/Doge-320M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SmallDoge/Doge-320M with Transformers:

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

pipe = pipeline("text-generation", model="SmallDoge/Doge-320M", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("SmallDoge/Doge-320M", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("SmallDoge/Doge-320M", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use SmallDoge/Doge-320M with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SmallDoge/Doge-320M"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SmallDoge/Doge-320M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SmallDoge/Doge-320M

SGLang

How to use SmallDoge/Doge-320M 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 "SmallDoge/Doge-320M" \
    --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": "SmallDoge/Doge-320M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "SmallDoge/Doge-320M" \
        --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": "SmallDoge/Doge-320M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use SmallDoge/Doge-320M with Docker Model Runner:
```
docker model run hf.co/SmallDoge/Doge-320M
```

JingzeShi commited on Aug 8, 2025

Commit

0cb1246

verified ·

1 Parent(s): 30f6869

Upload DogeForCausalLM

Browse files

Files changed (2) hide show

config.json +6 -6
model.safetensors +2 -2

config.json CHANGED Viewed

@@ -11,27 +11,27 @@
   "bos_token_id": 0,
   "dynamic_mask_ratio": 0.0,
   "eos_token_id": 1,
-  "expert_retrieval_size": 256,
   "hidden_act": "silu",
   "hidden_bias": false,
   "hidden_dropout": 0.0,
-  "hidden_size": 512,
   "initializer_range": 0.02,
-  "intermediate_size": 1024,
   "is_moe": false,
   "keep_window_size": 2048,
   "max_position_embeddings": 2048,
   "mlp_bias": false,
   "model_type": "doge",
   "norm_topk_prob": false,
-  "num_attention_heads": 4,
   "num_cdmoe_experts": 16348,
   "num_cdmoe_experts_per_head": 8,
   "num_cdmoe_heads": 4,
   "num_experts": 16384,
   "num_experts_per_tok": 64,
-  "num_hidden_layers": 16,
-  "num_key_value_heads": 2,
   "output_router_logits": false,
   "pad_token_id": 2,
   "rms_norm_eps": 1e-06,

   "bos_token_id": 0,
   "dynamic_mask_ratio": 0.0,
   "eos_token_id": 1,
+  "expert_retrieval_size": 64,
   "hidden_act": "silu",
   "hidden_bias": false,
   "hidden_dropout": 0.0,
+  "hidden_size": 768,
   "initializer_range": 0.02,
+  "intermediate_size": 1536,
   "is_moe": false,
   "keep_window_size": 2048,
   "max_position_embeddings": 2048,
   "mlp_bias": false,
   "model_type": "doge",
   "norm_topk_prob": false,
+  "num_attention_heads": 6,
   "num_cdmoe_experts": 16348,
   "num_cdmoe_experts_per_head": 8,
   "num_cdmoe_heads": 4,
   "num_experts": 16384,
   "num_experts_per_tok": 64,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 3,
   "output_router_logits": false,
   "pad_token_id": 2,
   "rms_norm_eps": 1e-06,

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0c0974eb9bdd4c39eb840024424b9ca67186461d7ad8a7b414289cedd3f2289d
-size 109174080

 version https://git-lfs.github.com/spec/v1
+oid sha256:c8ce781a449231f7744179cffa570d5f4e727f7867476b6810b9e8c8c5e80530
+size 305429944