Back to Models

Reflection 70B

mattshumer

Introduction

Reflection 70B, introduced on September 5, 2024, is an innovative AI model developed by HyperWrite in collaboration with Glaive AI. It utilizes Reflection-Tuning, a technique that enables the model to identify and correct its own errors during text generation. This feature improves both the accuracy and reliability of the outputs, making Reflection 70B suitable for academic research, content creation, and advanced problem-solving.

70B
100,000 tokens
150 tokens/s
16000 tokens/s

Chat

Input Tokens100,000 tokens
Output Tokens100,000 tokens
$0.0001/1K input, $0.0002/1K output

Release Information

Release Date2024-09-05
Versionreflection