Skip to content

DeepSeek V3.2

by DeepSeekRank #70Score 83.4

DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...

Performance Overview

Score
83.4
Rank
#70
24h Change
-1
7d Change
-6
State
stable
Confidence
high

Signal Scores

SignalNormalizedWeightContributionFreshness
Capabilities
capability
66.720%13.32026-09-22T14:23:19.816Z
Benchmarks
benchmark
86.730%26.02026-09-22T14:23:19.816Z
Pricing
pricing_tier
99.615%14.92026-09-22T14:23:19.816Z
Context Window
context_window
82.810%8.32026-09-22T14:23:19.816Z
Recency
recency
79.115%11.92026-09-22T14:23:19.816Z
Output Capacity
output_capacity
77.010%7.72026-09-22T14:23:19.816Z

Top Drivers

positive
Benchmarks
positive
Pricing
$0.40/M output tokens
$0.40
positive
Capabilities
Supports reasoning, tools, JSON mode, streaming
4/7
positive
Recency
Released over 1 year ago
10mo ago

Capabilities

CapabilitySupported
VisionNo
ReasoningYes
JSON ModeYes
StreamingYes
Function CallingYes
Web SearchNo

Pricing

Input / 1M tokens
$0.27
Output / 1M tokens
$0.40
Context Window
164K
Max Output
66K

Related Models

Related

DeepSeek V3.2 Tracker - Performance History | LM Market Cap