VERIFIEDOct 142026
Is DeepSeek V3.2 Exp Thinking state of the art?
No.Best current rank is #56 in Factuality.
As of Oct 14, 2026, DeepSeek V3.2 Exp Thinking from DeepSeek. Best current rank is #56 in Factuality. Ranked #56 in Factuality.
Why
The ruling, in plain English
Ranked #83 in General LLM.
Receipts
Evidence
| Source | Score | Date | Type | Receipt |
|---|---|---|---|---|
| LMArena Text (style-controlled, overall) deepseek-v3.2-exp-thinking | 1424.9010615201214 1418.3–1431.5 | 2026-10-13 | INDEP. | snap_7d2d37f2ecb4 |
| LMArena Text (style-controlled, hard prompts) deepseek-v3.2-exp-thinking | 1446.0575599361496 1437.5–1454.6 | 2026-10-13 | INDEP. | snap_7d2d37f2ecb4 |
| LMArena Text (style-controlled, math) deepseek-v3.2-exp-thinking | 1423.79639781867 1397.3–1450.3 | 2026-10-13 | INDEP. | snap_7d2d37f2ecb4 |
| LMArena Text (style-controlled, coding) deepseek-v3.2-exp-thinking | 1474.512470813816 1461.2–1487.8 | 2026-10-13 | INDEP. | snap_7d2d37f2ecb4 |
| LMArena Text (style-controlled, longer query) deepseek-v3.2-exp-thinking | 1434.001685398293 1421.4–1446.6 | 2026-10-13 | INDEP. | snap_7d2d37f2ecb4 |
| LMArena Text factuality deepseek-v3.2-exp-thinking | 1446.4372542693811 1435.7–1457.1 | 2026-10-13 | INDEP. | snap_af980e7d8ae9 |
Reign history
Where it sits in the succession
No closed or open crown interval includes this model.
Honest about the gaps
Where it is not the crown
| Category | This model | Why |
|---|---|---|
| General LLM | #83 in this category | Ranked #83 in General LLM. |
| Reasoning | #83 in this category | Ranked #83 in Reasoning. |
| Math | Not enough data | #85 on LMArena Text (style-controlled, math); missing Humanity's Last Exam and FrontierMath |
| Coding | Not enough data | #82 on LMArena Text (style-controlled, coding); missing Terminal-Bench 4.0 and SWE-bench Pro V2 HARD |
| Long context | #84 in this category | Ranked #84 in Long context. |
| Factuality | #56 in this category | Ranked #56 in Factuality. |