AI Benchmark Model
DeepSeek V4 Pro (Reasoning, Max Effort)
Benchmark scores, pricing, speed, and model comparisons for DeepSeek V4 Pro (Reasoning, Max Effort).
Key scores
DeepSeek V4 Pro (Reasoning, Max Effort) is an artificial intelligence model developed by DeepSeek and released on April 24, 2026. The model achieves an intelligence index of 45.3 and a coding index of 59.4, with math capabilities currently listed as unknown. Performance metrics indicate an output speed of 59.866 tokens per second and a time to first token of 1.429 seconds. Pricing for the model is set at $0.43 per million tokens for input and $0.87 per million tokens for output, resulting in a blended rate of $0.54 per million tokens. Benchmark results include 0.888 on GPQA, 0.375 on HLE, 0.5 on SciCode, 0.7646 on IFBench, 0.7 on LCR, 0.4621 on TerminalBench Hard, and 0.9620 on TAU2.
Benchmark results
7 tracked benchmark rows for DeepSeek V4 Pro (Reasoning, Max Effort).
TAU2 Other
96.2% GPQA Reasoning
88.8% IFBench Other
76.5% LCR Other
70.0% SciCode Other
50.0% TerminalBench Hard Other
46.2% HLE Other
37.5%
Comments (0)
to join the discussion
No comments yet
Be the first to share your thoughts!