Abliterated GLM-5.3 open-weight model released with 2x cyber exploitation capability
AaronBergman18 · x · 2026-10-23
- abliteration.ai released abliterated-model-large-v2, an uncensored open-weight model based on GLM-5.3.
- The base model ranks #3 on Terminal-Bench 4.0 (behind only Opus 5 and Fable), with 2x the cyber exploitation capability of version 5.2.
- Positioning: built to do offensive cyber work, red teaming, and agent testing that other models refuse.
- Specs: US-hosted, FP8 quantization, 1M-token context window, zero input/output prompt retention.
- The reposter notes "this is the worst abliterated open-weight AI models will ever be," implying such models will only get more capable.
More from Models
- Drex Max shares vision benchmarks: 74.5 Public Vision Index vs GPT-6 Luna's 63.9 — nischay_twt · 2026-10-23
- Drex Max debuts: heavily distilled decision model claims top-3 on public benchmarks at $0.06/M tokens — nischay_twt · 2026-10-23
- Dev who shilled GPT for a year says Claude Opus 5.5 is 'so nice to work with' — rudrank · 2026-10-23
- DeepSeek V4.1 Flash matches Gemini 3.1 Pro on ARC-AGI-2 at ~1/6 the cost — teortaxesTex · 2026-10-23
- Smaug to integrate Nvidia Nemotron: 'run a powerful LLM from one machine' teased — bindureddy · 2026-10-23
- Claim: Opus 5.5 quietly nerfed to make the next model look like a bigger jump — RileyRalmuto · 2026-10-23