DeepSeek

R1 Distill Llama 70B

by DeepSeektext->text1 endpoint
Neura Intelligence Index
44.7/ 100
Rank
#202
Confidence
low
Benchmarks
1(3% coverage)

Overview

DeepSeek R1 Distill Llama 70B is a distilled large language model based on Llama-3.3-70B-Instruct, using outputs from DeepSeek R1. The model combines advanced distillation techniques to achieve high performance across...

Capabilities

Text generation
Extended reasoning
Top-P sampling
Stop sequences
Deterministic seed

Modalities

Input
Text
Output
Text

Technical Specifications

Context Window
8.2K tokens
Max Output
7.4K tokens
Knowledge Cutoff
2024-07-31
Tokenizer
Llama3
Uptime (24h)
10000.0%
Parameters
70600000000B
License
mitOpen Weights
Country
🇨🇳CN
Release Date
January 20, 20251y

Supported Parameters

frequency_penaltyinclude_reasoningmax_tokenspresence_penaltyreasoningrepetition_penaltyseedstoptemperaturetop_ktop_p

Benchmark Performance

reasoning

BenchmarkScoreSource
GPQA Diamond
65.2
verified

Performance Over Time

Pricing

Per 1M tokens
Input$0.80
Output$0.80
Added
January 23, 2025
Last synced 9/20/2026