Back to Models

DeepSeek-V3.2 Reasoner

deprecated
DeepSeek2025-12-01
Specifications
Modalities
text
Context Window-
Max Output-
Training Cutoff-
Open SourceYes
Pricing (per 1M tokens)
Input$0.5500
Output$2.1900
Cache Read$0.1400
Benchmarks
Arena score (text)1422

Arena ratings: LMArena leaderboard dataset, CC BY 4.0, snapshot 2026-10-28.

MMLU-Pro-
SWE-Bench-
GPQA Diamond-
HumanEval-
ARC-AGI-
MATH-
API
Model IDdeepseek-reasoner