name = "DeepSeek-R1-Distill-Qwen-32B"
description = "R1 reasoning distilled into Qwen 2.5 32B for efficient open-weight step-by-step problem solving"
family = "deepseek-thinking"
release_date = "2025-01-20"
last_updated = "2025-01-20"
attachment = false
reasoning = true
temperature = true
tool_call = false
open_weights = true

[limit]
context = 131_072
output = 32_768

[modalities]
input = ["text"]
output = ["text"]

[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B"