DeepSeek V4 Pro (Reasoning, Max Effort)

DeepSeek

Frontier ModelReasoningMIT

Analysis of DeepSeek's DeepSeek V4 Pro (Reasoning, Max Effort) and comparison to other AI models across key metrics including quality, price, performance (tokens per second & time to first token), context window & more.

Context

Release Date: Apr 24, 2026
Window: 1M

PricingPer 1M tokens

Input: $0.435
Output: $0.87
Blended 3:1: $0.5438

Capabilities

Speed: 83 t/s
Input
Output
Reasoning tokens

Latency

TTFT: 1.79 ms
500 token response: 60.77 s

Benchmarks

Reasoning: ●●●○○
Coding: ●●○○○
GPQA: 88.8%
HLE: 35.9%
SciCode: 50.0%

Analysis of DeepSeek's DeepSeek V4 Pro (Reasoning, Max Effort) and comparison to other AI models across key metrics including quality, price, performance (tokens per second & time to first token), context window & more.