DeepSeek

DeepSeek V3.1

by DeepSeektext->text8 endpoints

Overview

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...

Capabilities

Text generation
Tool use / Function calling
Structured output
Extended reasoning
Prompt caching
Long context
Top-P sampling
Stop sequences
Deterministic seed
Log probabilities

Modalities

Input
Text
Output
Text

Technical Specifications

Context Window
163.8K tokens
Max Output
144.9K tokens
Knowledge Cutoff
2025-03-31
Tokenizer
DeepSeek
Uptime (24h)
9999.8%

Supported Parameters

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p

Pricing

Per 1M tokens
Input$0.55
Output$1.65
Cache read$0.55
Added
August 21, 2025
Last synced 8/27/2026