NVIDIA

DeepSeek V4 Pro 0813 — Benchmark Scores, Pricing & Performance Analysis

OPEN SOURCENVIDIA
Input Cost
$0.98/1M
Output Cost
$3.0/1M
Context Window
1.0M
Max Output
384K
Accepts
text

DeepSeek V4 Pro 0813 is an AI model by NVIDIA. View detailed benchmark scores, pricing data, and performance metrics on Serenities AI Models.

Specifications last checked against the provider on . Benchmark scores carry their own source, linked beside each number.

General Benchmarks

AA Intelligence Index
36.321st of 66

Artificial Analysis composite intelligence score across 10 sub-benchmarks

Artificial Analysis (via OpenRouter)

Coding Benchmarks

AA Coding Index
68.826th of 96

Artificial Analysis composite coding score

Artificial Analysis (via OpenRouter)

Reasoning Benchmarks

AA Agentic Index
42.316th of 70

Artificial Analysis composite score for multi-step tool-using tasks

Artificial Analysis (via OpenRouter)
GPQA Diamond
89.3%16th of 140

Graduate-level science Q&A by domain experts

Papers
τ²-Bench Airline
77.7%7th of 93

Multi-turn service agent making tool calls under strict policy constraints

OpenRouter (measured)

Cost Benchmarks

Input Cost
$0.98292nd of 413

Cost per 1M input tokens

OpenRouter API
Output Cost
$3.0279th of 413

Cost per 1M output tokens

OpenRouter API
Cached Input Cost
$0.0336th of 136

Cost per 1M cached input tokens — the price that actually applies to a long agent conversation

OpenRouter API

Context Benchmarks

Context Length
1.0M18th of 451

Maximum context window size

OpenRouter API

Available from 32 providers

The same model costs different amounts depending on who serves it. You can bring your own key for 8 of these — connect it here.

ProviderInput / 1MOutput / 1MCached inContextYour key
Alibaba Token Plan$0$0$01.0M
Alibaba Token Plan (China)$0$0$01.0M
Nvidia$0$01.0M
SCNet Token Plan$0$0$01.0M
OrcaRouter$0.442$0.884$0.061.0M
RunInfra$0.6$1.9$0.031.0M
Merge Gateway$0.66$1.98$0.0221.0M
Vercel AI Gateway$0.66$1.98$0.0661.0MSupported
AIHubMix$0.6918$2.0754$0.0230581.0M
OpenRouter$0.9834$2.9502$0.032781.0MSupported
NanoGPT$1.1$2.5$0.041.0M
Deep Infra$1.3$2.6$0.11.0MSupported
Weights & Biases$1.31$3.96$0.0441.0M
Arcee$1.32$3.96$0.0441.0M
Baseten$1.32$3.961.0MSupported
Cloudflare Workers AI$1.32$3.96$0.0441.0MSupported
DigitalOcean$1.32$3.96$0.0441.0M
Eden AI$1.32$3.96$0.131.0M
EmpirioLabs AI$1.32$3.96$1.321.0M
Fireworks AI$1.32$3.96$0.0441.0MSupported
Hugging Face$1.32$3.961.0MSupported
Kilo Gateway$1.32$3.96$0.0441.0M
Ofox$1.32$3.96$0.0441.0M
Together AI$1.32$3.96$0.131.0MSupported
Volcengine Ark$1.3359$4.00771$0.044531.0M
Vivgrid$1.35$3$0.051.0M
Charm Hyper$1.437216$4.311648$0.0479071.0M
Bothub$1.61$4.841.0M
Venice AI$1.65$4.95$0.1651.0M
Requesty$1.75$3.5$0.441.0M
Cortecs$2$3.999$0.51.0M
Ollama Cloud1.0M

DeepSeek V4 Pro 0813 — Benchmark Scores Overview

Scores normalized to percentage scale for visual comparison. ELO scores mapped to 0-100 range (1100-1500).

DeepSeek V4 Pro 0813 — Frequently Asked Questions

How much does DeepSeek V4 Pro 0813 cost?

DeepSeek V4 Pro 0813 costs $0.98 per 1M input tokens and $3.0 per 1M output tokens. This is mid-range pricing for its capability level.

What is the context window of DeepSeek V4 Pro 0813?

DeepSeek V4 Pro 0813 has a context window of 1.0M tokens. This determines how much text, conversation history, and code the model can process in a single request.

Who created DeepSeek V4 Pro 0813?

DeepSeek V4 Pro 0813 was created by NVIDIA. It is classified as a open source model in our catalogue.

Is DeepSeek V4 Pro 0813 open source?

Yes, DeepSeek V4 Pro 0813 is an open-source model. The model weights are publicly available for download and self-hosting.