DeepSeek V4 Pro 0813

DeepSeek · deepseek/deepseek-v4-pro-0813

← Back to leaderboard

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

DeepSeek V4 Pro 0813 ranks #8 of 348 models in the default AI Benchmark Hub order (quality first, then cost). At $2.24 blended per million tokens, it is cheaper than 33% of models with published API pricing. Its 1,024,000-token context window is 3.9× the catalog median (262,144 tokens). A quality index of 88.2 places it at #7 of 41 models with an Arena Elo signal.

open weightstexttext->text

Context

Max context: 1024000
Max output: 384000

Pricing

Input / 1M: 1.12
Output / 1M: 3.36
Blend / 1M: 2.24

Quality

Quality index: 88.2

Provider

Provider: DeepSeek
Moderated: no

Monthly cost at this blended rate

Estimated API spend if every token is billed at $2.24per million (the blend of this model's input and output prices). Real bills depend on the input/output mix.

Tokens / monthEst. cost (USD)
1,000,000$2.24
10,000,000$22.41
100,000,000$224.14

Related models

Same lab, similar price, or similar quality — useful next comparisons.

More from DeepSeek

Closest API price

Closest quality index