DeepSeek V3.2

Provider: DeepSeek CN

Input price
$0.28
/ 1M tokens
Output price
$0.42
/ 1M tokens
Cache price
$0.028
/ 1M tokens (read)
Context
164K tokens
Max output 147K tokens

Overview

DeepSeek V3.2 is a flagship reasoning model by DeepSeek targeting agentic tool use and high computational efficiency.

Strengths

  • Supports function-calling and json-mode
  • Features a 163,840 token context window
  • Includes a 147,456 token max output
  • Offers cached input priced at $0.028 per million

Best use cases

  • Building complex agentic tool pipelines
  • Processing structured JSON outputs at scale
  • Executing tasks requiring native thinking capabilities

Pricing position

Input sits at $0.28 and output at $0.42 per million tokens, making output roughly half the price of DeepSeek V3 0324 ($1.00) and Codestral 2508 ($0.90).

Caveats

There is no free tier available, and the context window is smaller than DeepSeek V4 Flash variants.

Notes drafted by an LLM from the figures on this page (2026-11-10) and rewritten when the price changes.

Key facts

Price basis
Median third-party host price (no first-party API)
Max output
147K tokens
Listed on OpenRouter
2025-12-01
Tracked since
2026-05-08
Last updated
2026-11-10

Features

Function callingJSON outputReasoning

Monthly cost estimate

Scale Input (M) Output (M) Monthly
0.1M (hobby) 0.1 0.05 $0.05
1M (small app) 1 0.5 $0.49
10M (mid scale) 10 5 $4.90
100M (large) 100 50 $49.00

※ Cache and batch discounts not factored in. Detailed calculator

Official resources

Other models from DeepSeek

Similar-priced models

Source: OpenRouter API (deepseek/deepseek-v3.2)