RLM idea: an 8B model reading n-gram tables, theoretically scaling to 1T params
gnukeith · x · 2026-11-05
gnukeith discusses RLM (Reading Language Model): an 8B model that reads n-gram tables, theoretically equivalent to a 1T-parameter model with only 8B active params — heavily caveated as theoretical. He also shares local deployment tips: running Qwen at very low quants degrades quality badly, but you can recover some of it using the DeepSeek Harness to find issues and fixing them with plugins/steering, plus links to small models that run at higher quants on a single GPU.
Related event: Dev shares low-end tricks for running Qwen3.8-Flash-Next and RLM concept(3 posts)→
More from Models
- $1.40 of uncensored GLM 5.3 + one scanner day finds 23 live leaked API keys — dbasch · 2026-11-05
- OpenAI's upcoming cyber model validated and patched 948 Linux privilege escalation vulnerabilities in one run — scaling01 · 2026-11-05
- Ricursive debuts from-scratch Ripple-65B: open-weight agentic coding SOTA at 1/10 cost — HazyResearch · 2026-11-05
- Sam Altman resurfaces ahead of rumored big OpenAI drop on Thursday — iruletheworldmo · 2026-11-05
- Epoch AI rebuilds its ECI with Bayesian methodology for sharper model comparisons — scaling01 · 2026-11-05
- Bittensor subnet to release ChronoLLM-16B, claiming lead in point-in-time financial LLMs — const_reborn · 2026-11-05