Model Details

• Base model: Qwen/Qwen3-30B-A3B (Mixture-of-Experts)
• Architecture: Qwen3MoeForCausalLM
• Total parameters: 30B (3B active per token, 8 experts routed per token)
• Hidden layers: 48
• Quantization: W7A7 (7-bit weights, 7-bit activations) via Pearl quantization
• Model type: Causal LLM (instruction-tuned, MoE)
• Primary runtime target: Pearl vLLM plugin (miner/vllm-miner)
• Intended use: Text generation with Pearl mining integration

Intended Use

This model is intended to be served through the Pearl miner stack, where vLLM inference is integrated with Pearl mining workflows.

Typical flow:

1. Run pearld with RPC enabled.
2. Start the Pearl miner/vLLM stack.
3. Serve this model through vLLM while Pearl gateway/miner components handle mining-side integration.

How To Use (Pearl vLLM Plugin)

Follow the miner setup from the Pearl repo:

• pearl/miner README

High-level prerequisites:

• Python 3.12
• uv
• CUDA + NVIDIA GPU (sm90 class, e.g. H100/H200, per project docs)
• Rust toolchain
• Running pearld node with RPC credentials

Docker Example

From the Pearl repository root:

bash
Copy code
docker buildx build -t vllm_miner . -f miner/vllm-miner/Dockerfile

bash
Copy code
docker run --rm -it --gpus all \
  -p 8000:8000 -p 8337:8337 -p 8339:8339 \
  -e PEARLD_RPC_URL=<PEARLD_URL> \
  -e PEARLD_RPC_USER=<RPC_USER> \
  -e PEARLD_RPC_PASSWORD=<RPC_PASSWORD> \
  -e VLLM_USE_DEEP_GEMM=0 \
  -v ~/.cache/huggingface:/root/.cache/huggingface \
  --shm-size 8g \
  vllm_miner:latest \
  pearl-ai/Qwen3-30B-A3B-Instruct-2507-pearl \
  --host 0.0.0.0 --port 8000 \
  --max-model-len 8192 \
  --gpu-memory-utilization 0.9 \
  --enforce-eager

vLLM Standalone

bash
Copy code
pip install vllm

vllm serve "pearl-ai/Qwen3-30B-A3B-Instruct-2507-pearl"

bash
Copy code
curl -X POST "http://localhost:8000/v1/chat/completions" \
  -H "Content-Type: application/json" \
  --data '{
    "model": "pearl-ai/Qwen3-30B-A3B-Instruct-2507-pearl",
    "messages": [
      {
        "role": "user",
        "content": "What is the capital of France?"
      }
    ]
  }'

SGLang

bash
Copy code
pip install sglang

python3 -m sglang.launch_server \
    --model-path "pearl-ai/Qwen3-30B-A3B-Instruct-2507-pearl" \
    --host 0.0.0.0 \
    --port 30000

bash
Copy code
curl -X POST "http://localhost:30000/v1/chat/completions" \
  -H "Content-Type: application/json" \
  --data '{
    "model": "pearl-ai/Qwen3-30B-A3B-Instruct-2507-pearl",
    "messages": [
      {
        "role": "user",
        "content": "What is the capital of France?"
      }
    ]
  }'

Transformers

python
Copy code
from transformers import pipeline

pipe = pipeline("text-generation", model="pearl-ai/Qwen3-30B-A3B-Instruct-2507-pearl")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

python
Copy code
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("pearl-ai/Qwen3-30B-A3B-Instruct-2507-pearl")
model = AutoModelForCausalLM.from_pretrained("pearl-ai/Qwen3-30B-A3B-Instruct-2507-pearl")

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]:]))

License

This model is based on Qwen3 and is distributed under the Apache 2.0 license.

Limitations

• Can generate incorrect, unsafe, or biased outputs.
• Requires careful deployment controls and output validation.
• Hardware/software compatibility depends on the Pearl miner stack and supported GPU architectures.