Skip to content
LLMs
DeepSeek logo

DeepSeek: DeepSeek V4 Pro 0813

by DeepSeek

DeepSeek: DeepSeek V4 Pro 0813 is a large language model from DeepSeek. It costs $0.983 per million input tokens and $2.95 per million output tokens. Its context window is 1.0M tokens.

ReasoningTool callingStructured outputPrompt cachingBatch tierOpen weights
Input / 1M tokens
$0.983
Output / 1M tokens
$2.95
Cached input / 1M
$0.033

On repeated prefixes

Context window
1.0M

tokens

Who serves it cheapest

20 hosts serve DeepSeek V4 Pro 0813. Same weights, same API - the price difference is pure margin and routing.

Providers serving DeepSeek: DeepSeek V4 Pro 0813, cheapest first
ProviderInput / 1MOutput / 1MContextThroughputUptime 24h
BaiduCheapestfp8$0.579$1.741.0M-99.9%
Alibaba$0.581$1.741M-99.9%
StreamLake$0.660$1.981.0M-99.6%
DeepSeek$0.660$1.981.0M-100.0%
Ionstream$0.960$2.881.0M-98.6%
Novitafp8$0.990$2.971.0M-99.9%
Phala$1.02$3.051.0M-99.7%
GMICloudfp8$1.06$3.171.0M-99.8%
DeepInfrafp8$1.30$2.601.0M-99.2%
NextBitfp8$1.12$3.371.0M-99.3%
CoreWeavefp8$1.31$3.961.0M-99.9%
Sail Researchfp4$1.32$3.961.0M-99.6%
BaseTenfp4$1.32$3.961.0M-99.0%
Parasailfp8$1.32$3.961.0M-97.8%
Together$1.32$3.961.0M-97.5%
DigitalOcean$1.32$3.961.0M-99.9%
SiliconFlowfp8$1.32$3.961.0M-96.2%
Cloudflare$1.32$3.961.0M-95.4%
Fireworks$1.32$3.961.0M-99.8%
Venice$1.65$4.951M-98.0%

The spread between Baidu and Venice is 2.8× for identical weights. Quantization and context limits differ, so check both columns before switching.

Benchmarks

Independent scores published alongside the catalogue.

Intelligence index
36.3
Coding index
68.8
Agentic index
42.3

About 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

DeepSeek: DeepSeek V4 Pro 0813 specifications
Model IDdeepseek/deepseek-v4-pro-0813
ProviderDeepSeek
Context window1.0M tokens
Max output384K tokens
Input modalitiestext
Output modalitiestext
Knowledge cutoff-
Open weightsYes - deepseek-ai/DeepSeek-V4-Pro-0813
ReleasedAugust 12, 2026

Cheaper alternatives

Models that cost less than DeepSeek V4 Pro 0813 while keeping at least half its context window and every input modality it supports.

Qwen3 Coder Plus

$0.650 in · $3.25 out

1.1× cheaper

Hy4 preview

$0.834 in · $2.50 out

1.2× cheaper

Gemini 3 Flash Preview

$0.500 in · $3.00 out

1.3× cheaper

GLM 5.2

$0.683 in · $2.15 out

1.4× cheaper

MiniMax M1

$0.550 in · $2.20 out

1.5× cheaper

Frequently asked

How much does DeepSeek: DeepSeek V4 Pro 0813 cost?

$0.983 per million input tokens and $2.95 per million output tokens. Cached input reads cost $0.033 per million tokens.

What is the context window of DeepSeek: DeepSeek V4 Pro 0813?

1.0M tokens, with up to 384K tokens of output per request.

Which provider serves DeepSeek: DeepSeek V4 Pro 0813 cheapest?

Baidu at $0.579 per million input tokens - 2.8× cheaper than Venice, the most expensive of the 20 hosts serving it.

Confirm against the source: DeepSeek official pricing.