# Is DeepSeek V3.2 Exp SOTA?

**It used to be.** Was SOTA in Reasoning.

As of Oct 14, 2026, DeepSeek V3.2 Exp from DeepSeek. Was SOTA in Reasoning. Was SOTA in Long context from 2025-11-09 to 2025-11-17. Currently ranked #80.

As of Oct 14, 2026. Methodology v1.0. Computed automatically.

- General LLM: RANKED #90 — Ranked #90 in General LLM.
- Reasoning: RANKED #80 — Ranked #80 in Reasoning.
- Math: INSUFFICIENT_DATA — #100 on LMArena Text (style-controlled, math); missing Humanity's Last Exam and FrontierMath
- Coding: INSUFFICIENT_DATA — #92 on LMArena Text (style-controlled, coding); missing Terminal-Bench 4.0 and SWE-bench Pro V2 HARD
- Coding › Web dev: RANKED #96 — Ranked #96 in Coding › Web dev.
- Long context: WAS_SOTA #80 — Was SOTA in Long context from 2025-11-09 to 2025-11-17. Currently ranked #80.
- Factuality: RANKED #73 — Ranked #73 in Factuality.