Instructions to use CharyZeng/DeepSeek-V4-Flash-4layer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CharyZeng/DeepSeek-V4-Flash-4layer with Transformers:

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

pipe = pipeline("text-generation", model="CharyZeng/DeepSeek-V4-Flash-4layer")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("CharyZeng/DeepSeek-V4-Flash-4layer")
model = AutoModelForCausalLM.from_pretrained("CharyZeng/DeepSeek-V4-Flash-4layer")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use CharyZeng/DeepSeek-V4-Flash-4layer with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CharyZeng/DeepSeek-V4-Flash-4layer"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CharyZeng/DeepSeek-V4-Flash-4layer",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/CharyZeng/DeepSeek-V4-Flash-4layer

SGLang

How to use CharyZeng/DeepSeek-V4-Flash-4layer 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 "CharyZeng/DeepSeek-V4-Flash-4layer" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CharyZeng/DeepSeek-V4-Flash-4layer",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "CharyZeng/DeepSeek-V4-Flash-4layer" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CharyZeng/DeepSeek-V4-Flash-4layer",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use CharyZeng/DeepSeek-V4-Flash-4layer with Docker Model Runner:
```
docker model run hf.co/CharyZeng/DeepSeek-V4-Flash-4layer
```

CharyZeng commited on 12 days ago

Commit

b773bea

verified ·

1 Parent(s): d9556ce

4-layer extract: hidden layers 0,1,2 + MTP

Browse files

Files changed (4) hide show

config.json +1 -41
model-00001-of-00002.safetensors +3 -0
model-00002-of-00002.safetensors +3 -0
model.safetensors.index.json +0 -0

config.json CHANGED Viewed

@@ -24,7 +24,7 @@
   "norm_topk_prob": true,
   "num_attention_heads": 64,
   "num_experts_per_tok": 6,
-  "num_hidden_layers": 4,
   "num_hash_layers": 3,
   "num_key_value_heads": 1,
   "num_nextn_predict_layers": 1,
@@ -66,46 +66,6 @@
     0,
     0,
     4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
-    128,
-    4,
     0
   ]
 }

   "norm_topk_prob": true,
   "num_attention_heads": 64,
   "num_experts_per_tok": 6,
+  "num_hidden_layers": 3,
   "num_hash_layers": 3,
   "num_key_value_heads": 1,
   "num_nextn_predict_layers": 1,
     0,
     0,
     4,
     0
   ]
 }

model-00001-of-00002.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:b49b08cb592a5f481d69113cc7f098437e09e4803addb92f5a1006f065f5d2c1
+size 10734101852

model-00002-of-00002.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:2f51b7462d4cac23b3e369b88fe7a1d3acfbb4ef6a1ba4ce07de6e6b0d8f6983
+size 5707127668

model.safetensors.index.json CHANGED Viewed

The diff for this file is too large to render. See raw diff