PipeLLM logo
Back to approved models

Approved model profile_

DeepSeekactive

Deepseek V4 Pro

DeepSeek V4 Pro is a large-scale Mixture-of-Experts model aimed at advanced reasoning, coding, long-context analysis, and long-horizon agent workflows. It is suited for full-codebase analysis, multi-step automation, and large-scale information synthesis where capability and efficiency both matter.

deepseek-v4-pro
Open Relay View integration
relay/request.mjsready to route
const completion = await client.chat.completions.create({
  model: "deepseek-v4-pro",
  messages: [
    { role: "user",
      content: "Plan a multi-step task" }
  ]
});
policy checked provider selected
CONTEXT1M contextprompt window
MAX OUTPUT384Kper response
INPUT$1.3 / M30% OFF · list $1.8 / M
OUTPUT$2.5 / M30% OFF · list $3.6 / M

Route configuration_

Use the model through the contract you already have.

PipeLLM resolves provider access, policy, and supported model capabilities without changing the interface your application calls.

ENDPOINThttps://api.pipellm.aiOne base URL for OpenAI, Anthropic, and Gemini.

FORMAT ROUTES

OPENAI/openai
ANTHROPIC/anthropic
GEMINI/gemini
Use a format route to normalize requests and responses.
MODEL IDdeepseek-v4-proUse this in every request
GOVERNANCEPolicy controlledApprovals and tool rules apply
PROVIDER AVAILABILITY1 route
ProviderContextMax outputDiscountInput / MOutput / M
Volcano Engine1M context384K30% OFF$1.3 / M$2.5 / M
Displayed pricing is per one million tokens and may vary by provider.
SUPPORTED SURFACESactive
input: textoutput: texttool usecache controlcomputer use

Relay governs model access with the same routing policy that Lens keeps visible and auditable.