Is DeepSeek V4 Flash Preview SOTA in Factuality?#101 in FactualityRanked #101 in Factuality.
How it scores here
DeepSeek V4 Flash Preview in Factuality
Primary benchmarks
- LMArena Text factuality: 1434 (CI 1430–1437) on Nov 19, 2026 (#120 of 215).
On the category's weighted score it is 59.4 points behind #1 Gemini 4 Carbon and 0.6 behind the next model up, Mistral Large 3.
Nearby on the board
| # | Model | Category score |
|---|---|---|
| 98 | Claude Opus 4.1 | 42.4 |
| 99 | Grok 4.1 | 41.8 |
| 100 | Mistral Large 3 | 41.2 |
| 101 | DeepSeek V4 Flash Preview | 40.6 |
| 102 | Gemini 3.1 Flash Lite Preview | 40.0 |
| 103 | Minimax M3 | 39.4 |
| 104 | Nvidia Nemotron 3 Ultra 550B A55b | 38.8 |
Standings
Factuality
| # | Model | State |
|---|---|---|
| 1 | Gemini 4 Carbon | Co-SOTA |
| 1 | Gemini 4 Neon | Co-SOTA |
| 2 | Claude Opus 5.5 | Near-SOTA |
| 3 | Claude Fable 5.1 | Near-SOTA |
| 4 | Muse Wonder | Near-SOTA |
| 5 | Gemini 4 Hydrogen | Near-SOTA |
| 6 | Claude Opus 4.6 | Near-SOTA |
| 7 | Gemini 4 Argon | Was SOTA |
| 8 | Claude Opus 5 | Was SOTA |
| 9 | Mimo V2.6 Pro | #9 in this category |
| 10 | GPT-5.5 | Was SOTA |
| 11 | GPT-5.4 | Was SOTA |