DeepSeek V3.1
by DeepSeekRank #106Score 71.8
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...
Performance Overview
Score
71.8
Rank
#106
24h Change
0
7d Change
+258
State
stableConfidence
highSignal Scores
| Signal | Normalized | Weight | Contribution | Freshness |
|---|---|---|---|---|
Capabilities capability | 66.7 | 30% | 20.0 | 2026-11-06T14:58:51.103Z |
Pricing pricing_tier | 99.1 | 25% | 24.8 | 2026-11-06T14:58:51.103Z |
Context Window context_window | 82.8 | 15% | 12.4 | 2026-11-06T14:58:51.103Z |
Recency recency | 52.3 | 15% | 7.9 | 2026-11-06T14:58:51.103Z |
Output Capacity output_capacity | 82.6 | 15% | 12.4 | 2026-11-06T14:58:51.103Z |
Top Drivers
positive
Pricing
$0.95/M output tokens
$0.95
positive
Capabilities
Supports reasoning, tools, JSON mode, streaming
4/7
positive
Context Window
164K token context window
164K
positive
Output Capacity
Up to 147K output tokens per request
147K
Capabilities
| Capability | Supported |
|---|---|
| Vision | No |
| Reasoning | Yes |
| JSON Mode | Yes |
| Streaming | Yes |
| Function Calling | Yes |
| Web Search | No |
Pricing
Input / 1M tokens
$0.25
Output / 1M tokens
$0.95
Context Window
164K
Max Output
147K