The Orchestrator › Models › DeepSeek

DeepSeek: DeepSeek V4 Pro 0813

DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. This is the GA release of DeepSeek V4 Pro.

Specifications

ProviderDeepSeek
Model IDdeepseek/deepseek-v4-pro-0813
Context window1.0M tokens (1,048,576)
Modalitiestext → text
TokenizerDeepSeek
Router sourcesOpenRouter, TrustedRouter, Vercel AI Gateway, Cloudflare
Released2026-08-12
Tracked since2026-11-14
PopularityNot in the ranked set
ReasoningSupported

Pricing (per million tokens)

VariantSourceAPI IDInputOutputContext
standardOpenRouterdeepseek/deepseek-v4-pro-0813$0.111$4.441.0M
standardTrustedRouterdeepseek/deepseek-v4-pro-0813$1.39$4.181.0M
standardVercel AI Gatewaydeepseek/deepseek-v4-pro-0813$0.66$1.981M
standardCloudflaredeepseek/deepseek-v4-pro-0813$1.32$3.96—

Recorded price history

DateInput / 1MOutput / 1MContext
2026-11-14$1.32$3.96—
2026-11-14$0.66$1.981M
2026-11-14$1.39$4.181.0M
2026-11-14$0.111$4.441.0M

Capabilities

Agentic tool useStructured outputs1M+ context windowControllable reasoningNo content storage (attested)Zero data retention

Trust flags come from TrustedRouter's per-provider attestations.

Supported parameters

frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p