name = "DeepSeek: DeepSeek V4 Pro"
description = "Flagship DeepSeek model for coding, reasoning, and agentic work"
release_date = "2026-04-24"
last_updated = "2026-05-01"
attachment = false
reasoning_options = [{ type = "toggle" }, { type = "effort", values = ["high", "xhigh"] }]
reasoning = true
tool_call = true
temperature = true
open_weights = false

[cost]
input = 0.435
output = 0.87
cache_read = 0.003625

[limit]
context = 1048576
output = 384000

[modalities]
input = ["text"]
output = ["text"]