DeepSeek 3.2

by DeepSeekChatunverified match

DeepSeek chat model for instruction following, coding, and analysis

Context
164K tokens
Max output
163,840 tokens
Released
2025-12-02
Weights
Open

deepseek/deepseek-3-2

Specifications

Full facts as recorded, incl. what is unknown.

Released
2025-12-02
Weights
Open
Input
Text
Output
Text
Context window
163,840 max across endpoints
Max output
163,840
Providers
1
Epoch Capabilities Index
not evaluated
Direct price range
$0.50 in · $1.60 out per 1M tokens · from its one provider
More details
Last updated
2026-04-30
Knowledge cutoff
2024-12

Capabilities

Across all endpoints: supported if any endpoint supports it.

  • Tool calling
  • Parallel toolsunknown
  • Structured outputunknown
  • Reasoning
  • Image input
  • PDF input
  • Audio input
  • Audio outputunknown
  • Video input
  • Prompt cachingunknown
  • Web searchunknown
  • Computer useunknown
  • System messagesunknown
  • Assistant prefillunknown
  • Temperature
  • Streamingunknown
  • Attachments

Endpoints

One row per provider and variant; gateways are listed separately.

Loading endpoints…

Price history

No change since 19 Oct 2026. The cheapest direct price has been $0.50 in / $1.60 out per 1M tokens at DigitalOcean on every data refresh since then.

Tracked from 19 Oct 2026. Headline-eligible direct prices only, so a price the sanity check flags never draws a spike. Series as JSON

Cost calculator

Loading endpoints…

More from DeepSeek

Newest DeepSeek models in the catalogue. All DeepSeek models

ModelReleasedContextListed price
DeepSeek V4 Flash2026-10-171.05Mfrom $0.07
DeepSeek V4.1 Flash 10172026-10-171Mfrom $0.08
DeepSeek V4 Flash Vision Exp2026-10-171.05Mfrom $0.14
DeepSeek V4.1 Flash2026-09-101.05Mfrom $0.09
DeepSeek V4 Pro 08132026-08-121Mfrom $0.88
DeepSeek V4 Flash Latest2026-07-311.05Mfrom $0.04
DeepSeek Pro Latest2026-07-061Munpriced
DeepSeek Latest2026-05-031.05M$0.145 via NanoGPT
Heads up: prices are indicative list rates scraped from provider pricing pages; verify with the provider before relying on them. Report incorrect data for this model