from encoding_dsv4 import encode_messages, parse_message_from_completion_text

messages = [
    {"role": "user", "content": "hello"},
    {"role": "assistant", "content": "Hello! I am DeepSeek.", "reasoning_content": "thinking..."},
    {"role": "user", "content": "1+1=?"}
]

# messages -> string
prompt = encode_messages(messages, thinking_mode="thinking", reasoning_effort="max")

# string -> tokens
import transformers
tokenizer = transformers.AutoTokenizer.from_pretrained("deepseek-ai/DeepSeek-V4-Flash-0731")
tokens = tokenizer.encode(prompt)

如何使用 vLLM 运行

只需添加一个标志即可启用 DSpark 推测解码——在您的 vLLM 启动命令中添加 --speculative-config，并指定方法为 dspark：

--speculative-config '{"method":"dspark","num_speculative_tokens":7,"draft_sample_method":"greedy"}'

例如，以下命令在单个 4×GB300 节点上使用 vLLM 部署模型。 有关详细说明和其他硬件配置，请参见 vLLM 指南。

vllm serve deepseek-ai/DeepSeek-V4-Flash-0731 
  --trust-remote-code --kv-cache-dtype fp8 --block-size 256 
  --data-parallel-size 4 --enable-expert-parallel 
  --moe-backend deep_gemm_mega_moe 
  --attention-config '{"use_fp4_indexer_cache": true}' 
  --speculative-config '{"method":"dspark","num_speculative_tokens":7,"draft_sample_method":"greedy"}'

sglang serve 
  --trust-remote-code 
  --model-path deepseek-ai/DeepSeek-V4-Flash-0731 
  --tp 4 
  --moe-runner-backend flashinfer_mxfp4 
  --speculative-algorithm DSPARK 
  --mem-fraction-static 0.90 
  --chunked-prefill-size 4096 
  --swa-full-tokens-ratio 0.1 

@misc{deepseekai2026deepseekv4,
      title={DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence},
      author={DeepSeek-AI},
      year={2026},
}