返回资讯中心
外部精选
软件工程
#开发者实时流

DeepSeek 在 Hugging Face 发布 DeepSeek-V4-Pro-0813 模型

来源标题:DeepSeek V4 Pro 0813

DeepSeek 在 Hugging Face 上线了 DeepSeek-V4-Pro-0813,模型卡提供 Transformers、vLLM 等库的调用示例,可直接加载推理。该模型采用 fp8 精度,许可证为 mit,并附有 arXiv 论文 2606.19348 与评测结果。

Hacker News 最新nthypes2026/11/14 · 4 天前30 分钟阅读

以下正文同步自 Hacker News 最新,版权归原站所有,已转换为易读排版。

Like 915

Follow

DeepSeek 154k

Text Generation

Transformers

Safetensors

deepseek_v4

Eval Results

8-bit precision

fp8

arxiv: 2606.19348

License: mit

Model card Files Files and versions xet

Community 25

Deploy

Copy to bucket

Use this model Instructions to use deepseek-ai/DeepSeek-V4-Pro-0813 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

Transformers How to use deepseek-ai/DeepSeek-V4-Pro-0813 with Transformers:

Use a pipeline as a high-level helper

from transformers import pipeline

pipe = pipeline("text-generation", model="deepseek-ai/DeepSeek-V4-Pro-0813")

pip install -U transformers accelerate

Load model directly

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("deepseek-ai/DeepSeek-V4-Pro-0813") model = AutoModelForCausalLM.from_pretrained("deepseek-ai/DeepSeek-V4-Pro-0813", device_map="auto")

Inference

Inference Providers

HuggingChat

Notebooks

Google Colab

Kaggle

Local Apps Settings

vLLM How to use deepseek-ai/DeepSeek-V4-Pro-0813 with vLLM:

Install vLLM from pip:

pip install vllm

Start the vLLM server:

vllm serve "deepseek-ai/DeepSeek-V4-Pro-0813"

Call the server using curl (OpenAI-compatible API):

curl -X POST "http://localhost:8000/v1/completions"
-H "Content-Type: application/json"
--data '{ "model": "deepseek-ai/DeepSeek-V4-Pro-0813", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'

Use Docker

docker model run hf.co/deepseek-ai/DeepSeek-V4-Pro-0813

SGLang How to use deepseek-ai/DeepSeek-V4-Pro-0813 with SGLang:

Install SGLang from pip:

pip install sglang

Start the SGLang server:

python3 -m sglang.launch_server
--model-path "deepseek-ai/DeepSeek-V4-Pro-0813"
--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": "deepseek-ai/DeepSeek-V4-Pro-0813", "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 "deepseek-ai/DeepSeek-V4-Pro-0813"
--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": "deepseek-ai/DeepSeek-V4-Pro-0813", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'

Docker Model Runner How to use deepseek-ai/DeepSeek-V4-Pro-0813 with Docker Model Runner:

docker model run hf.co/deepseek-ai/DeepSeek-V4-Pro-0813

Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

DeepSeek-V4-Pro-0813 Introduction

Chat Template

How to Run with vLLM

How to Run with SGLang

How to Run Locally

License

Citation

Contact

DeepSeek-V4-Pro-0813

Technical Report👁️

Introduction

DeepSeek-V4-Pro-0813 is the official release of DeepSeek-V4-Pro, superseding the preview version, with greatly enhanced agentic capabilities and performance improvements that are especially pronounced in production environments. It is built on the DeepSeek-V4-Pro (Preview) model structure, with a DSpark speculative decoding module attached.

DeepSeek-V4-Pro-0813 outperforms DeepSeek-V4-Pro (Preview) on the benchmarks listed below, and is broadly competitive with the strongest proprietary models available.

Chat Template

This release does not include a Jinja-format chat template. Instead, we provide a dedicated encoding folder with Python scripts and test cases demonstrating how to encode messages in OpenAI-compatible format into input strings for the model, and how to parse the model's text output. Please refer to the encoding folder for full documentation.

The reasoning_effort parameter now supports three levels — low, high, and max — which control how much deliberation the model spends before answering.

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-Pro-0813") tokens = tokenizer.encode(prompt)

How to Run with vLLM

DSpark speculative decoding is enabled with a single flag — add --speculative-config with method: dspark to your vLLM launch command:

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

For example, the command below serves the model with vLLM on a single 4×GB300 node. See the vLLM recipe for detailed instructions and other hardware configurations.

vllm serve deepseek-ai/DeepSeek-V4-Pro-0813
--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"}'

How to Run with SGLang

Enable DSpark with --speculative-algorithm DSPARK and do not set a separate --speculative-draft-model-path as the target and draft weights therefore come from the same checkpoint. See the SGLang cookbook for detailed instructions, benchmarks and other hardwares configurations.

sglang serve
--trust-remote-code
--model-path deepseek-ai/DeepSeek-V4-Pro-0813
--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 \

How to Run Locally

Please refer to the inference folder for detailed instructions on running DeepSeek-V4 locally, including model weight conversion and interactive chat demos.

For local deployment, we recommend setting the sampling parameters to temperature = 1.0, with top_p = 0.95 for agentic scenarios and top_p = 1.0 otherwise. For the high and max reasoning effort levels, we recommend a maximum output length of 384K tokens.

License

This repository and the model weights are licensed under the MIT License.

Citation

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

Contact

If you have any questions, please raise an issue or contact us at service@deepseek.com.

","lstrip":false,"normalized":true,"rstrip":false,"single_word":false},"eos_token":{"__type":"AddedToken","content":"<|end▁of▁sentence|>","lstrip":false,"normalized":true,"rstrip":false,"single_word":false},"unk_token":null,"pad_token":{"__type":"AddedToken","content":"<|end▁of▁sentence|>","lstrip":false,"normalized":true,"rstrip":false,"single_word":false}}},"createdAt":"2026-08-13T03:05:06.000Z","discussionsDisabled":false,"discussionsSorting":"recently-created","downloads":77882,"downloadsAllTime":409976,"id":"deepseek-ai/DeepSeek-V4-Pro-0813","isLikedByUser":false,"availableInferenceProviders":[{"provider":"together","modelStatus":"live","providerStatus":"live","providerId":"deepseek-ai/DeepSeek-V4-Pro-0813","task":"conversational","adapterWeightsPath":"model-00001-of-00066.safetensors","features":{"structuredOutput":false,"toolCalling":true},"isCheapestPricingOutput":false,"isFastestThroughput":false,"isModelAuthor":false,"tokensPerSecond":61.766578313597606,"pricingOutput":3.96},{"provider":"novita","modelStatus":"live","providerStatus":"live","providerId":"deepseek/deepseek-v4-pro-0813","task":"conversational","adapterWeightsPath":"model-00001-of-00066.safetensors","features":{"structuredOutput":true,"toolCalling":true},"isCheapestPricingOutput":false,"isFastestThroughput":false,"isModelAuthor":false,"tokensPerSecond":74.9109407748759,"pricingOutput":3.96},{"provider":"deepinfra","modelStatus":"live","providerStatus":"live","providerId":"deepseek-ai/DeepSeek-V4-Pro-0813","task":"conversational","adapterWeightsPath":"model-00001-of-00066.safetensors","features":{"structuredOutput":true,"toolCalling":true},"isCheapestPricingOutput":true,"isFastestThroughput":false,"isModelAuthor":false,"tokensPerSecond":31.081448177262338,"pricingOutput":2.5999999999999996},{"provider":"baseten","modelStatus":"live","providerStatus":"live","providerId":"deepseek-ai/DeepSeek-V4-Pro-0813","task":"conversational","adapterWeightsPath":"model-00001-of-00066.safetensors","features":{"structuredOutput":false,"toolCalling":true},"isCheapestPricingOutput":false,"isFastestThroughput":true,"isModelAuthor":false,"tokensPerSecond":84.82406159749216,"pricingOutput":3.96}],"showHuggingChatEntry":true,"inference":"warm","lastModified":"2026-08-13T16:28:28.000Z","likes":915,"pipeline_tag":"text-generation","library_name":"transformers","librariesOther":[],"trackDownloads":true,"model-index":null,"evalResults":[{"dataset":{"id":"harborframework/terminal-bench-2.1","isBenchmark":true,"task_id":"terminalbench_2_1"},"value":87.9,"source":{"url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813","name":"Model Card","isExternal":false},"filename":".eval_results/terminal-bench-2.1.yaml","verified":false,"pullRequest":19,"rank":4,"label":"Terminalbench 2 1","notes":"DeepSeek Harness (minimal mode), max reasoning effort, temperature=1.0, top_p=0.95."},{"dataset":{"id":"datacurve/deep-swe","isBenchmark":true,"task_id":"deep_swe"},"value":62.7,"source":{"url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813","name":"Model Card","isExternal":false},"filename":".eval_results/deep-swe.yaml","verified":false,"pullRequest":25,"rank":16,"label":"Deep Swe","notes":"DeepSWE v1.1."},{"dataset":{"id":"hkust-nlp/Toolathlon","isBenchmark":false,"task_id":"toolathlon_verified"},"value":74.1,"source":{"url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813","name":"deepseek-ai/DeepSeek-V4-Pro-0813 model card","isExternal":false},"filename":".eval_results/deepseek-ai__DeepSeek-V4-Pro-0813.yaml","verified":false,"pullRequest":21,"rank":5,"label":"Toolathlon Verified"}],"private":false,"repoType":"model","gated":false,"tags":["transformers","safetensors","deepseek_v4","text-generation","arxiv:2606.19348","license:mit","eval-results","endpoints_compatible","8-bit","fp8","region:us"],"tag_objs":[{"id":"text-generation","label":"Text Generation","type":"pipeline_tag","subType":"nlp"},{"id":"transformers","label":"Transformers","type":"library"},{"id":"safetensors","label":"Safetensors","type":"library"},{"id":"deepseek_v4","label":"deepseek_v4","type":"other","clickable":true},{"id":"eval-results","label":"Eval Results","type":"other","clickable":true},{"id":"endpoints_compatible","label":"Inference Endpoints","type":"other","clickable":true},{"id":"8-bit","label":"8-bit precision","type":"other","clickable":true},{"id":"fp8","label":"fp8","type":"other","clickable":true},{"id":"arxiv:2606.19348","label":"arxiv:2606.19348","type":"arxiv","extra":{"paperTitle":"DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence"}},{"id":"license:mit","label":"mit","type":"license"},{"type":"region","label":"🇺🇸 Region: US","id":"region:us"}],"transformersInfo":{"auto_model":"AutoModelForCausalLM","pipeline_tag":"text-generation","processor":"AutoTokenizer"},"widgetData":[{"text":"Hi, what can you help me with?"},{"text":"What is 84 * 3 / 2?"},{"text":"Tell me an interesting

正文由 探How 从来源站点 RSS 同步,内容未经改写;遇到排版缺失或需要图片、视频时请以原文为准。

打开原文
去社区讨论 →