Instructions to use 0xSero/DeepSeek-V3.2-345B-W3A16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use 0xSero/DeepSeek-V3.2-345B-W3A16 with Transformers:

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

pipe = pipeline("text-generation", model="0xSero/DeepSeek-V3.2-345B-W3A16")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("0xSero/DeepSeek-V3.2-345B-W3A16")
model = AutoModelForCausalLM.from_pretrained("0xSero/DeepSeek-V3.2-345B-W3A16")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use 0xSero/DeepSeek-V3.2-345B-W3A16 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "0xSero/DeepSeek-V3.2-345B-W3A16"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "0xSero/DeepSeek-V3.2-345B-W3A16",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/0xSero/DeepSeek-V3.2-345B-W3A16

SGLang

How to use 0xSero/DeepSeek-V3.2-345B-W3A16 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 "0xSero/DeepSeek-V3.2-345B-W3A16" \
    --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": "0xSero/DeepSeek-V3.2-345B-W3A16",
		"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 "0xSero/DeepSeek-V3.2-345B-W3A16" \
        --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": "0xSero/DeepSeek-V3.2-345B-W3A16",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use 0xSero/DeepSeek-V3.2-345B-W3A16 with Docker Model Runner:
```
docker model run hf.co/0xSero/DeepSeek-V3.2-345B-W3A16
```

Support this work → · X · GitHub · REAP paper · Cerebras REAP

DeepSeek-V3.2-345B-W3A16

W3A16 quantization of deepseek-ai/DeepSeek-V3.

At a glance


Base model	deepseek-ai/DeepSeek-V3
Format	W3A16
Total params	345B
Active / token	—
Experts / layer	128
Layers	61
Hidden size	7168
Context	163,840
On-disk size	138 GB

Which variant should I pick?

Variant	Format	Link
`DeepSeek-V3.2-345B-W3A16` (this)	W3A16	link
`DeepSeek-V3.2-508B-NVFP4`	NVFP4	link

𓌳 REAP𓌳 the Experts: Why Pruning Prevails for One-Shot MoE Compression
📄 Paper • 💻 Code

DeepSeek-V3.2-REAP-345B-W3A16

REAP-pruned + W3A16 quantized DeepSeek-V3.2 for efficient deployment.

📋 Model Specifications

Property	Value
Base Model	DeepSeek-V3.2
Parameters	345B
Quantization	W3A16 (3-bit weights)

🔬 Calibration Dataset: Deep Dive

REAP's effectiveness depends critically on calibration data that represents the target use case. We specifically optimized for code generation, function/tool calling, and agentic workflows.

Why These 3 Datasets?

Dataset	Samples	Purpose	Why It Matters
evol-codealpaca-v1	700	Code generation	51% of mix — Code tasks activate specific expert pathways; pruning without code calibration destroys coding ability
xlam-function-calling-60k	330	Function/tool calling	24% of mix — Tool use requires structured JSON output; experts handling schema generation must be preserved
SWE-smith-trajectories	330	Agentic multi-turn	24% of mix — Real SWE-bench trajectories with tool calls, file edits, and multi-step reasoning

The Science Behind Dataset Selection

REAP Algorithm:
1. Forward pass calibration samples through model
2. Record which experts activate and their magnitudes
3. Compute saliency = router_weight × activation_norm
4. Prune lowest-saliency experts

Key Insight: Experts are TASK-SPECIFIC
├── Some experts specialize in natural language
├── Some experts specialize in code syntax
├── Some experts specialize in JSON/structured output
└── Some experts specialize in multi-turn context

If calibration lacks code → code-specialized experts appear "unused" → get pruned → model loses coding ability

Cerebras' Original Mix (from paper)

Cerebras used the same 3 datasets in their GLM-4.6 REAP experiments:

evol-codealpaca-v1 for code generation
xlam-function-calling-60k for tool calling
SWE-smith-trajectories for agentic tasks

We followed this exact recipe for reproducibility.

Combined Dataset

Our calibration mix: 0xSero/glm47-reap-calibration-v2

License & citation

License inherited from the base model.

@misc{lasby2025reap,
  title  = {REAP the Experts: Why Pruning Prevails for One-Shot MoE Compression},
  author = {Mike Lasby and Ivan Lazarevich and Nish Sinnadurai and Sean Lie and Yani Ioannou and Vithursan Thangarasa},
  year   = {2025}, eprint = {2510.13999}, archivePrefix = {arXiv}
}