Nesformer: an entire Nintendo crammed into ~1GB of Qwen3.5 weights, following the Qwenputer method
idanbeck · x · 2026-10-31
Divyansh's Qwenputer embedded a full RISC-V CPU inside Qwen3.5-4B weights, using the model's own Gated DeltaNet layers as RAM, running C and MicroPython in forward passes with no external runtime. On DeepMind math: base 55–90%, base + Python tool 95–100%, Qwenputer 97.5–100%.
Drew Breunig applied the same approach to build Nesformer, fitting an entire Nintendo console into roughly 1GB of Qwen3.5 weights.
More from Research
- NVIDIA Nemotron 3.5 scales LLM self-verification, +9-10 points on agentic benchmarks — sarahcat21 · 2026-10-31
- Unofficial OpenTargets MCP Server Brings Gene-Drug-Disease Data to LLMs — modelcontextprotocol · 2026-10-31
- Researchers find 5 personality structures in LLM weight space, Big Five-like — luke_drago_ · 2026-10-31
- NVS benchmarks are misaligned: proper pose alignment puts 3D Gaussians back at SOTA — kwangmoo_yi · 2026-10-31
- arXiv moderator: submission surge causes months-long holds, new moderators wanted — tdietterich · 2026-10-31
- arXiv policy checks reject class projects, proposals, surveys and position papers — tdietterich · 2026-10-31