Find tools › Open-source projects › Research the web

qwenputer

Open-source project · by divagr18

This experimental tool builds a virtual mini-computer directly into an AI model's internal memory to run exact calculations.

For developers

What it costs

Free and open source; you may pay for the AI model it uses.

Use it to

  • Run C and MicroPython code directly inside model forward passes
  • Execute exact multi-digit math without external software sandboxes
  • Research compiled virtual hardware built inside neural network weights

How to start

  1. Clone the repository and install dependencies using git clone https://github.com/divagr18/qwenputer && cd qwenputer/experimental/riscv && uv sync --extra hf.

  2. Download the compiled model weights using hf download divagr1925/Qwen3.5-4B-Qwenputer --local-dir qwenputer-rv.

  3. Run an arithmetic query using uv run python -m chipformer.rv.chat qwenputer-rv --firmware tests/data/rv/calc-bridge.elf --ask "Use calc: what is 123456789 * 987654321?".

Good to know. This is an experimental research project requiring substantial graphics hardware (like an NVIDIA L4 GPU) to execute effectively. The repository has no specified license.

Before you install

  • Found automaticallyNot reviewed by a person yet. Check it before installing.
  • Last updated 2026-11-09

More detail

Qwenputer compiles a 32-bit RISC-V CPU directly into the weights of the Qwen3.5-4B language model.

It enables exact arithmetic and deterministic program execution directly inside a model's forward pass without relying on external code sandboxes. Reach for it when researching neural network interpretability or compiled weight architectures.

Similar tools

Descriptions are written in plain English from each project's own documentation and refreshed by our agent; this one was last checked 2026-11-08. Spotted a mistake? · Open in Find tools · © 2026 Which AI Map