Tech · GitHub
Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on LiveCodeBench Hard
Compiled by KHAO Editorial — aggregated from 1 source. See llms.txt for citation guidance.
◌ Single Source
A multi-agent harness for programming problems, written as a Memdoor workflow.
Key facts
- A run took a median of 40 minutes of model time (the longest 4 hours) and $0.39 — It is the method of GVS5H (Lee, 2026: five Qwen3.8-27B models match Fable 5 on LiveCodeBench Hard)
- The coder wrote these files itself from a description of the method (2026-10-25)
- The contests date from 29 March to 6 April 2025, and the model may have seen them in training: contamination cannot be ruled out
Summary
Rebuilt with the authors' exact prompts on Qwen3.8-27B, their loop scored a mean of 9.8 of 10 over five passes on the first ten LiveCodeBench Hard problems, against 9.6 for their own harness and 8.0 for Fable 5 alone ( below, every run published). The coder wrote these files itself from a description of the method (2026-10-25). GVS5H's own harness, rebuilt as a Memdoor workflow on the same model: Qwen3.8-27B, a 128K reply cap, the authors' prompts and loop. The authors' columns are their repository's regraded result files ( q38_multiagent_pN, fable5_single_pN ), the numbers in their paper. bench/results/parity-5pass/ holds all 50 runs: every request and reply, the program, the grader's full output, each call's cost and the host that served it, and every attempt that did not count.