DeepSeek: DeepSeek V3.1
deepseek/deepseek-chat-v3.1
Context: 163,840In: $0.25/1MOut: $0.95/1MUpdated 2026-08-05
DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It extends the DeepSeek-V3 base with a two-phase long-context...
Intelligence
—
Coding
—
Agentic
—
Metric profile
Trend over time
Intelligence
—
Coding
—
Agentic
—
Only 1 snapshot so far (data refreshes daily) — the trend line appears after the 2nd snapshot.
Classic evals (curated from model cards)
DeepSeek — manually maintained, may lag releases
MMLU-Pro
75%
GPQA Diamond
67%
MATH-500
95%
Design Arena (ELO by category)
| Category | ELO | Win rate | Rank |
|---|---|---|---|
| website | 1146 | 48% | #77 |
| codecategories | 1141 | 47.9% | #77 |
| 3d | 1133 | 48% | #72 |
| gamedev | 1129 | 47.1% | #74 |
| dataviz | 1126 | 46.8% | #76 |
| uicomponent | 1120 | 47.6% | #74 |
| svg | 1014 | 38.2% | #75 |