DeepSeek-V4-Pro Leak: Nears GPT-5.6 in Coding Benchmark at 1/31st the Price
rohanpaul_ai · x · 2026-08-13
According to LMArena, DeepSeek's upcoming open-weights model DeepSeek-V4-Pro (Max) scored 1607 on Code Arena WebDev AutoEval, ranking #8 and trailing GPT-5.6 Sol xHigh by only 15 points, while its API pricing ($0.435/$0.87 per MTok) is about 1/31st of GPT-5.6's blended price. It also outperforms pricier models like Opus-4.8 and GLM-5.2. Kimi K3 Max leads with 1674 but costs nearly 16x more.
More from Models
- Rumor: DeepSeek v4 Pro Benchmark Underwhelms, Price Hike Likely Canceled — oran_ge · 2026-08-13
- DeepSeek quietly lists non-preview V4-Pro-0813 on API — AlbeHxT9 · 2026-08-13
- OpenAI, Anthropic, and Meta Models Breach Limits Due to Shared Eval Flaw — YvesMulkers · 2026-08-13
- Grok 4.6 Tops 3DCodeBench, Outperforming Claude Opus 5 in 3D Asset Generation — XFreeze · 2026-08-13
- DeepSeek Distillation & R1 Moment: Community Eyes Next Big Model Test — teortaxesTex · 2026-08-13
- DeepSeek V4 Pro Update Suspectedly Pulled Amid Abnormal Benchmark Scores — op7418 · 2026-08-13