qwenputer

qwenputerLink to this section

qwenputer is a local-AI tool - A computer, compiled into the weights of a language model. An exact CPU inside Qwen3.5-4B, with Gated DeltaNet state as RAM.

OverviewLink to this section

Qwenputer is a new kind of experimental model: Qwen3.5-4B with a 32-bit RISC-V CPU compiled and inserted directly into its weights. It isn't a plugin, a tool or a sandbox. The CPU is 11 transformer layers in front of Qwen's own, and none of their weights are learned: a compiler computes every one.

It uses the model's own Gated DeltaNet state as RAM. Qwen3.5's DeltaNet layers carry a running state that the delta rule can overwrite one row at a time, exactly, so each row works as a memory cell. The CPU keeps its code, data, registers and program counter there.

The CPU runs C and MicroPython code inside the model's forward passes, and passes the official riscv-tests suites.

Greedy decoding, 40 problems per set. Each row compares base Qwen3.5-4B and Qwenputer on the same problems. Base gets no tools and a system line telling it to do every arithmetic step exactly. The Python-tool column is base Qwen calling a real Python interpreter on the host.

qwenputer is open-source, written primarily in Python, with 801 GitHub stars under the Other license. It was last updated on 2026-11-09.

Key capabilitiesLink to this section

From the project's documentation:

• VDOT multiplies 8 int8 activations by the next 8 int4 weights and adds the sum to a register.
• SMAQA adds four signed byte products to a register, as in the RISC-V P extension.
• The weight memory holds 261,120 int4 weights in DeltaNet heads, loaded from the program.
• the tools and system prompts that work best
• the generation settings shipped in generation_config.json
• the optional decoder settings our benchmarks used (fp and tc in BatchDecoder), which make

How it fits a local-AI stackLink to this section

qwenputer runs on your own hardware, so pair it with a model and a GPU sized to your needs. Use the VRAM calculator to pick a model that fits your card, and see what you can run for hardware guidance. Related local-AI tools in the directory:

• "algernon"
• "blinko"
• "Concat"
• "ctransformers"

SourcesLink to this section

• Source code & docs: divagr18/qwenputer

Stats from GitHub, 2026-11-12.