Qwen3.8 27B sums big numbers in words, hitting 167/169 with reasoning on
Simon Willison re-ran an experiment Colin Frasier first tried two years ago with GPT-4o: have a model compute large sums but return the answer in words. He ran it on local hardware (a DGX Spark), using a Codex Remote session (GPT-6 Astra) to write and execute the test with Qwen3.8-27B-Q4_K_M.gguf. With reasoning disabled he sampled 30 attempts per cell; with reasoning enabled each run took so long that he used one sample per pair, and the model got 167 of 169 attempts right.
Why it matters: Digit-by-digit addition in prose is often treated as a stress test of model arithmetic, and the result suggests a local 27B-class model can handle it fairly reliably once reasoning is enabled.
Research: Qwen3.8 27B addition in words Colin Frasier posted on Bluesky about an experiment he ran over two years ago using GPT-4o to see how well it could "compute the sum but return the answer in words" across increasingly large numbers. Here's the chart he shared of those results: I'm confident GPT-4o didn't cheat and use a calculator, especially since it got so many of the calculations wrong, but I was inspired to run the experiment again on local hardware (a DGX Spark) to explore the effect in a fully controlled environment. I pasted his image into a Codex Remote session (GPT-6 Astra) and had it run the same experiment using Qwen3.8-27B-Q4_K_M.gguf.
Here's the result for a run of 30 attempts per combination with reasoning disabled: Then I ran it again with reasoning enabled. This took a lot longer per pair, so instead of running 30 samples per square I ran just one - which results in a much less visually appealing heatmap since each square is either 100% or 0%: It got the right answer in 167 out of 169 attempts, and since these were one-shot I'm confident a second run would produce different results here. Here's a version of the report that includes the reasoning traces from some of those larger calculations, which include text like this: Wait, let me redo this more carefully.
4,299,366,105,622 6,088,794,067,970 Let me align them: 4 2 9 9 3 6 6 1 0 5 6 2 2 6 0 8 8 7 9 4 0 6 7 9 7 0 Adding from right to left: Position 1 (units): 2 + 0 = 2 Position 2 (tens): 2 + 7 = 9 Position 3 (hundreds): 6 + 9 = 15, write 5, carry 1 Tags: mathematics, ai, generative-ai, local-llms, llms, qwen, llm-reasoning, dgx-spark