DeepSeek · Released Apr 14, 2026
DeepSeek V4 Pro 0813 (max)
DeepSeek's cutting-edge open-weights model delivering exceptional coding, reasoning, and mathematical performance at minimal cost.
OpenRating Index
53.4
0–100 composite
Reasoning Index
75.2
0–100 composite
Cost / Task
$0.25
benchmark run
Output speed
82 tok/s
blended tok/s
Time to first token
0.7s
median latency
Input price
$0.22
per 1M tokens
Output price
$0.88
per 1M tokens
Context window
256K
tokens
Strengths
Outstanding cost-to-performance ratio ($0.25 per task)
Unmatched open-weights coding benchmark scores
Permissive license with active community ecosystem
How we rate DeepSeek V4 Pro 0813 (max)
OpenRating scores every model on the same axes. The OpenRating Index is a weighted composite of reasoning, knowledge, and agentic benchmarks, normalised to a 0–100 scale. Speed is blended output measured in tokens per second under standardised load, with time-to-first-token reported separately. Prices are list input and output prices per million tokens, and Task Cost measures end-to-end workload spending.
OpenRating Index
53.4
Reasoning Index
75.2
Token Pricing
Input per 1M tokens
$0.22
Output per 1M tokens
$0.88
Cost of 1M input + 1M output
$1.1
Task Cost Economics
Empirical cost breakdown for a standard OpenRating Index benchmark task ($0.25 total).
Related models
Compare DeepSeek V4 Pro 0813 (max) with the models it is most often measured against.