richardyoung/ deepseek-r1-0528-qwen3-8b-heretic

191 2 days ago

Uncensored DeepSeek-R1-0528-Qwen3-8B via Heretic: 4/100 refusals (from 89), KL 0.041. Reasoning model, MIT.

5 models

deepseek-r1-0528-qwen3-8b-heretic:latest
256f71265a04 • 5.0GB • 128K context window •
Text input • 2 days ago
deepseek-r1-0528-qwen3-8b-heretic:Q4_K_M
256f71265a04 • 5.0GB • 128K context window •
Text input • 2 days ago
deepseek-r1-0528-qwen3-8b-heretic:Q5_K_M
1c1d59bff4ae • 5.9GB • 128K context window •
Text input • 2 days ago
deepseek-r1-0528-qwen3-8b-heretic:Q6_K
9ecf74fbe235 • 6.7GB • 128K context window •
Text input • 2 days ago
deepseek-r1-0528-qwen3-8b-heretic:Q8_0
c6ab0a0d9fd3 • 8.7GB • 128K context window •
Text input • 2 days ago