reflection

105.5K 8 months ago

A high-performing model trained with a new technique called Reflection-tuning that teaches a LLM to detect mistakes in its reasoning and correct course.

70b

17 models

reflection:latest
5084e77c1e10 • 40GB • 128K context window •
Text input • 8 months ago
reflection:70b latest
5084e77c1e10 • 40GB • 128K context window •
Text input • 8 months ago
reflection:70b-q2_K
8fe3c853372c • 26GB • 128K context window •
Text input • 8 months ago
reflection:70b-q3_K_S
21f651100031 • 31GB • 128K context window •
Text input • 8 months ago
reflection:70b-q3_K_M
a6b22bd90923 • 34GB • 128K context window •
Text input • 8 months ago
reflection:70b-q3_K_L
9c6705916e06 • 37GB • 128K context window •
Text input • 8 months ago
reflection:70b-q4_0
5084e77c1e10 • 40GB • 128K context window •
Text input • 8 months ago
reflection:70b-q4_1
b72afde19a06 • 44GB • 128K context window •
Text input • 8 months ago
reflection:70b-q4_K_S
420791ca0c2a • 40GB • 128K context window •
Text input • 8 months ago
reflection:70b-q4_K_M
be39ad6154f4 • 43GB • 128K context window •
Text input • 8 months ago
reflection:70b-q5_0
99e430b53c8b • 49GB • 128K context window •
Text input • 8 months ago
reflection:70b-q5_1
41bd1db0b708 • 53GB • 128K context window •
Text input • 8 months ago
reflection:70b-q5_K_S
84a4d89b332c • 49GB • 128K context window •
Text input • 8 months ago
reflection:70b-q5_K_M
f537d644476a • 50GB • 128K context window •
Text input • 8 months ago
reflection:70b-q6_K
77fecce26024 • 58GB • 128K context window •
Text input • 8 months ago
reflection:70b-q8_0
159e9e593c44 • 75GB • 128K context window •
Text input • 8 months ago
reflection:70b-fp16
e04ae4d96458 • 141GB • 128K context window •
Text input • 8 months ago