AIToday
Large Language ModelsOpen-Source AILobsters AIPublished: Oct 29, 2026, 22:00 JST

Qwenputer puts a working RISC-V CPU inside Qwen 3.5 4B weights

Send on LINE
Qwenputer puts a working RISC-V CPU inside Qwen 3.5 4B weights

3 Key Points

  1. What happened

    Divyansh Agrawal released Qwenputer, a stock Qwen 3.5 4B with a 32-bit RISC-V CPU compiled into its weights across 11 extra layers, using Gated DeltaNet state as RAM. It passes the official riscv-tests and has run 18 million instructions free-running without a single error.

  2. Why it matters

    That points to models that can execute exact computations internally rather than relying on outside tools, though Agrawal presents it as an experiment.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Context & Analysis

Qwenputer builds on a property of Qwen 3.5's architecture. Three out of four attention layers in the model use Gated DeltaNet, a linear RNN, and each head keeps a state matrix. Agrawal found that by setting the gate and write strength to 1 and using one-hot key encoding, those state matrices can act as exact, addressable RAM, at lower cost than using normal attention because a GDN head's state stays the same size however long the conversation gets.

With that RAM in place, no training was needed: a compiler sets every weight directly, turning CPU steps into attention heads that look things up, feed-forward units that do the arithmetic, and DeltaNet heads that hold the memory. The CPU then runs as the model decodes, with each instruction becoming a few generated tokens that form the execution trace.

Agrawal also extended the idea to storage. A filesystem kept in the DeltaNet state stores up to 512 files, about 53,000 tokens of text, and persists when state is saved and loaded into a new conversation. He frames exact control over the RNN layers as a testbed for safety research, including whether hybrid models can hide reasoning in their RNN state and whether compiled weights can serve as monitors the model cannot switch off.

FAQ
Does Qwenputer need an external tool server or Python sandbox to do calculations?
No. The CPU runs arithmetic, C and MicroPython code inside the model's forward passes, with no external runtime or tool usage whatsoever.
What hardware do I need to try Qwenputer?
It runs on a single 24 GB GPU. You also need uv and the Hugging Face CLI, and the code is on GitHub and the model on Hugging Face and ModelScope.
Is the CPU's output exact?
Agrawal replayed about 230,000 of the CPU's output positions against a reference machine and found zero mismatches.
Send on LINE

Also reported by GIGAZINE AI

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleNscale seeks $30 billion IPO as investors push back to $16B-$19B