Leaked DeepSeek V4 Pro Benchmarks Show It Beating Claude Opus in Multiple Metrics
dotey · x · 2026-08-13
A user shared recent evaluation details regarding the deepseek-v4-pro model on social media. According to the quoted benchmark screenshots, the model has surpassed Claude 3 Opus in several metrics. This suggests DeepSeek may have internally tested or quietly released a powerful new version.
More from Models
- DeepSeek V4 Pro Update Suspectedly Pulled Amid Abnormal Benchmark Scores — op7418 · 2026-08-13
- Qwen3.8-27B Coming Soon: Release Set for August 14 on Hugging Face — huggingface · 2026-08-13
- DeepSeek-V4-Pro Leak: Nears GPT-5.6 in Coding Benchmark at 1/31st the Price — rohanpaul_ai · 2026-08-13
- DeepSeek V4 API Fingerprint Changes, Hinting at New Checkpoints — teortaxesTex · 2026-08-13
- Users Report Severe Model Degradation Across Google's APIs — EthanBeMe · 2026-08-13
- Qwen3.8-27B Model Surfaces on ModelScope Ahead of Hype — Ok-Shower7286 · 2026-08-13