Skip to content

DeepSeek V3.1

by DeepSeekRank #151Score 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
#151
24h Change
+1
7d Change
-45
State
stable
Confidence
high

Signal Scores

SignalNormalizedWeightContributionFreshness
Capabilities
capability
66.720%13.32026-08-08T12:27:18.102Z
Benchmarks
benchmark
69.130%20.72026-08-08T12:27:18.102Z
Pricing
pricing_tier
99.115%14.92026-08-08T12:27:18.102Z
Context Window
context_window
82.810%8.32026-08-08T12:27:18.102Z
Recency
recency
68.715%10.32026-08-08T12:27:18.102Z
Output Capacity
output_capacity
75.010%7.52026-08-08T12:27:18.102Z

Top Drivers

positive
Benchmarks
positive
Pricing
$0.95/M output tokens
$0.95
positive
Capabilities
Supports reasoning, tools, JSON mode, streaming
4/7
positive
Recency
Released over 1 year ago
12mo ago

Capabilities

CapabilitySupported
VisionNo
ReasoningYes
JSON ModeYes
StreamingYes
Function CallingYes
Web SearchNo

Pricing

Input / 1M tokens
$0.25
Output / 1M tokens
$0.95
Context Window
164K
Max Output
33K

Related Models

Related

DeepSeek V3.1 Tracker - Performance History | LM Market Cap