Benchmark27 Aug 2026 00:00 UTC · 33d ago

Epoch AI evaluates DeepSeek v4 Pro (high)

SimpleQA Verified 47.0%

Type: BENCHMARK_RESULT
Detection: benchmark
Confidence: high
Status

Related entities

ModelDeepSeek v4 Pro (high)CompanyDeepSeek

Primary source

Epoch AI Benchmarking Hub · logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F ↗

Independent benchmark runs (GPQA Diamond, SWE-bench Verified, FrontierMath, ...) plus externally reported leaderboards and the Epoch Capabilities Index (ECI).

Related events

Open release5mo ago
DeepSeek releases DeepSeek v4 Pro (high)
Epoch AI Benchmarking HubDeepSeek v4 Pro (high)DeepSeek