Abliterated GLM 5.3 goes viral, but the technique is 3 years old
thursdai_pod · x · 2026-09-12
ThursdAI podcaster @petergostev's release of an abliterated GLM 5.3 for offensive cybersecurity went viral, but the abliteration technique itself is roughly 3 years old. The episode also covers abliteration.ai, an OpenAI/Anthropic-compatible unrestricted LLM API for red teams, trust & safety, synthetic data, and research — drop-in via baseURL change, with training-data generation and schema-validated outputs.
More from Models
- DeepSeek v4.1 Flash runs out of the box on six NVIDIA GPUs via vLLM on day 0, AMD lags — woosuk_k · 2026-09-12
- ARC-AGI cost collapse: DeepSeek-V4-Flash hits 87% at $0.021/task vs o3's $4,500 — inductionheads · 2026-09-12
- Open uncensored AI has demand, but distribution has been stuck for two years — Rumbleblak · 2026-09-12
- DeepSeek V4.1 Flash cuts global KV cache to 890 bytes/token, but HBM demand may rise with agent swarms — teortaxesTex · 2026-09-12
- Claude user complains of contradictory usage limits: unlock time vs 5-hour reset confusion — XCxBigDong69XCx · 2026-09-12
- SGLang hits 873 tok/s on DeepSeek V4.1 Flash within 24 hours of launch — BanghuaZ · 2026-09-12