DeepSeek-R1-Distill-Llama-70B

Parameters

70B

Published

Jan 22, 2025

DeepSeek-R1-Distill-Llama-70B is an AI model developed by DeepSeek (China), first published in January 2025. It works in the language domain, on tasks such as language modeling/generation, question answering, quantitative reasoning and code generation.

Epoch AI has no training-compute estimate for this model. The model has 70,000,000,000 parameters. It was trained on roughly 24B datapoints.

Access: Open weights (unrestricted). Its weights are openly available. It is built on top of Llama 3.3 70B. Epoch AI rates the confidence of this record as speculative.

Full record

Organization: DeepSeek
Country of organization: China
Domain: Language
Task: Language modeling/generation, Question answering, Quantitative reasoning, Code generation
Compute estimation method: Operation counting
Parameters: 70,000,000,000
Dataset size: 24B
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Base model: Llama 3.3 70B
Epoch confidence: Speculative

Benchmark results

01	LiveBench	Global average	54.5
02	MATH Level 5	mean_score	0.9
03	GPQA Diamond	mean_score	0.56
04	OTIS Mock AIME 2024-2025	mean_score	0.51