Skip to content
RankShark

DeepSeek V4 Pro 0813

DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek.

ModelOpen sourceby DeepSeeksince 2026
36.9

SharkScore

#172 in Models

Why it scores 36.9

DeepSeek V4 Pro 0813 has a SharkScore of 36.9/100, ranked #172 of 608 in Models (updated 2026-11-19).

Each part compares DeepSeek V4 Pro 0813 with other listings in Models. Evidence covers 87% of the score's weight.

Adoption
11 · 30%
User reviews
no data · 25%
Maintenance
37 · 20%
Trust & transparency
80 · 15%
Momentum
no data · 10%

Paying never changes this score. How ranking works · method v2026-11-18.2

RankShark review of DeepSeek V4 Pro 0813

Automated evaluation by RankShark · 18 Nov 2026 · method v1.0 · How we review

We were not paid for this review.

As of 2026-11-18, DeepSeek V4 Pro 0813 is a text-to-text model with a 1,048,576-token context window; OpenRouter lists usage pricing of $0.111 per million input tokens and $4.44 per million output tokens. [E2]

OpenRouter describes DeepSeek V4 Pro 0813 as DeepSeek's large-scale mixture-of-experts model and general-availability release. The model is text-to-text and supports a context length of 1,048,576 tokens. [E2]

Best for

  • Developers evaluating a model for coding or agentic workflows. · model card, OpenRouter
  • Workloads that can use a long text context. · OpenRouter
  • Teams that want to run the model with vLLM or SGLang. · model card

Not for

  • Workloads requiring image or audio input or output; OpenRouter lists text as the only input and output modality. · OpenRouter

Key features

  • Long context. OpenRouter lists a context length of 1,048,576 tokens. · OpenRouter
  • Reasoning effort controls. The vendor says reasoning effort supports low, high, and max. OpenRouter also lists none as a supported effort. · model card, OpenRouter
  • Agentic and coding benchmarks. The vendor reports benchmark results including Terminal Bench 2.1, DeepSWE, Toolathlon-Verified, and AutomationBench. · model card
  • Inference options. The vendor provides instructions for running the model with vLLM and SGLang, including DSpark speculative decoding configuration. · model card
  • MIT license. The repository lists the model weights and repository under the MIT License. · model card
  • Message encoding. The vendor says the release does not include a Jinja-format chat template and provides Python scripts and test cases for encoding and parsing messages. · model card

Pros

  • OpenRouter lists a 1,048,576-token context window. · OpenRouter
  • The model listing supports multiple reasoning effort settings. · OpenRouter
  • The vendor provides instructions for local deployment with vLLM and SGLang. · model card
  • The repository lists the model weights under the MIT License. · model card

Cons

  • OpenRouter lists text as the only input and output modality. · OpenRouter
  • The vendor says the release does not include a Jinja-format chat template; users are directed to its encoding scripts and documentation. · model card
  • No free plan or free trial stated on the pages we checked.
  • Whether customer data is used for model training is not stated on the pages we checked.

Pricing

PlanPriceLimits
OpenRouter listing$0.111 per million input tokens; $4.44 per million output tokensOpenRouter lists a context length of 1,048,576 tokens.

Checked 18 Nov 2026

Free plan
Not stated (checked 18 Nov 2026)
Free trial
Not stated (checked 18 Nov 2026)
Trains on customer data
Not stated (checked 18 Nov 2026)
Data retention
Not stated (checked 18 Nov 2026)
Data regions
Not stated (checked 18 Nov 2026)
Data processing agreement
Not stated (checked 18 Nov 2026)
AI models used
DeepSeek V4 Pro 0813 · model card, OpenRouter
API
OpenRouter lists chat completions, completions, messages, and responses endpoints. · OpenRouter
SDKs
Not stated (checked 18 Nov 2026)
Webhooks
Not stated (checked 18 Nov 2026)
Platforms
The vendor provides model files and instructions for local use, including deployment with vLLM and SGLang. · model card
Security claims
No security certification stated (checked 18 Nov 2026)

What we measured

Context window (OpenRouter listing) · now
1,048,576 tokens
Price per million tokens (OpenRouter listing) · now
$0.111 in, $4.44 out
Package downloads · 30 days
46,938
How we tested · sources
  • Hourly uptime and response-time checks of the product's URL.
  • Repository, release and download data from GitHub, npm, PyPI and Hugging Face where they exist.
  • We did not sign up, pay, or use the product as a customer.
  • 18 Nov 2026: Plans or prices changed on the pricing page.

Something wrong? Report an error with a public link that shows it; corrections re-run the review.

Top alternatives to DeepSeek V4 Pro 0813

See all
  • 67.1

    Qwen3.8 27B is an open-weight dense vision-language model from Qwen.

    ModelOpen source
  • 65.5

    Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per t

    ModelOpen source
  • 64.7

    Qwen3.8 Flash is a multimodal reasoning model from Alibaba.

    ModelOpen source
  • 64.3

    Qwen3.6 27B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026.

    ModelOpen source