Instructions to use hmellor/tiny-random-DeepseekV2ForCausalLM with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hmellor/tiny-random-DeepseekV2ForCausalLM with Transformers:

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

pipe = pipeline("text-generation", model="hmellor/tiny-random-DeepseekV2ForCausalLM")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("hmellor/tiny-random-DeepseekV2ForCausalLM")
model = AutoModelForCausalLM.from_pretrained("hmellor/tiny-random-DeepseekV2ForCausalLM", device_map="auto")
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 Settings

vLLM

How to use hmellor/tiny-random-DeepseekV2ForCausalLM with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "hmellor/tiny-random-DeepseekV2ForCausalLM"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hmellor/tiny-random-DeepseekV2ForCausalLM",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/hmellor/tiny-random-DeepseekV2ForCausalLM

SGLang

How to use hmellor/tiny-random-DeepseekV2ForCausalLM 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 "hmellor/tiny-random-DeepseekV2ForCausalLM" \
    --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": "hmellor/tiny-random-DeepseekV2ForCausalLM",
		"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 "hmellor/tiny-random-DeepseekV2ForCausalLM" \
        --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": "hmellor/tiny-random-DeepseekV2ForCausalLM",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use hmellor/tiny-random-DeepseekV2ForCausalLM with Docker Model Runner:
```
docker model run hf.co/hmellor/tiny-random-DeepseekV2ForCausalLM
```

tiny-random-DeepseekV2ForCausalLM

Randomly initialised DeepseekV2ForCausalLM for testing MLA (multi-head latent attention) and MoE code paths without loading a 29 GiB checkpoint. Used by vLLM's tests/models/transformers/test_backend.py::test_mla.

The attention head dimensions are copied verbatim from deepseek-ai/DeepSeek-V2-Lite-Chat so that kv_lora_rank + qk_rope_head_dim == 576, the latent head size every MLA kernel is built around. Only the width, depth and expert counts are shrunk.

	DeepSeek-V2-Lite-Chat	this model
`kv_lora_rank`	512	512
`qk_nope_head_dim`	128	128
`qk_rope_head_dim`	64	64
`v_head_dim`	128	128
`hidden_size`	2048	256
`num_hidden_layers`	27	4
`num_attention_heads`	16	4
`n_routed_experts`	64	8
`n_shared_experts`	2	2
parameters	15.7B	59.8M

Layer 0 is dense and the rest are MoE (first_k_dense_replace=1), matching the real model. The tokenizer is DeepSeek-V2-Lite-Chat's, unmodified.

The weights are random, so the outputs are meaningless. This is only useful for checking that two implementations agree with each other.

Downloads last month: -

Safetensors

Model size

59.8M params

Tensor type

BF16

Collection including hmellor/tiny-random-DeepseekV2ForCausalLM

vLLM

Collection

Models used in vLLM CI • 5 items • Updated about 9 hours ago • 2