Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use

Details

FieldValue
iddeepseek-v3-2
familydeepseek
modalitytext->text
context_length128000
max_output_tokens32768
knowledge_cutoff2024-07
release_date2025-12-01
is_open_weightsyes
supports_tool_callyes
supports_reasoningyes
supports_structured_outputyes
supports_attachmentno
input_modalitiestext
output_modalitiestext

Providers

DeepSeek V3.2