"Living Weights" ships in TensorFold: local models update their own weights while in use
EAccelerate_42 · x · 2026-10-10
Developer ashxhart released "Living Weights," three months of engineering now shipping in the latest TensorFold release with support for NVIDIA Nemotron Lightning. The idea challenges the norm that open models ship frozen: the model keeps learning and updating its own weights during use, without KV stashing or MD-file tricks.
Selling points: the model becomes uniquely yours, and unloading it to another machine preserves what it learned. The author frames it as a possible start of recursive self-improvement for local AI, though technical details remain thin.
More from Models
- Test shows Haiku 5.5 triples robot task success with more thinking, GPT-6 Luna stays near zero — ycombinator · 2026-10-10
- Open-EmbeddingGemma thread wraps up: why Kye Gomez reimplements closed-source models — KyeGomezB · 2026-10-10
- Open-EmbeddingGemma: pure PyTorch reimplementation of Google's 740M multimodal embedding model — KyeGomezB · 2026-10-10
- Sources behind the Open-EmbeddingGemma reimplementation: HF model page and Google blog — KyeGomezB · 2026-10-10
- How EmbeddingGemma 2 works: six steps from five modalities to one 768-dim vector — KyeGomezB · 2026-10-10
- Dev claims Gemini 3.8 Flash beats Opus 5.5 for traditional RAG in finance — rickasaurus · 2026-10-10