← Chen, Ho Yiing — Research Records

Bragi-LLM: An 805 MB Hybrid Code-Generation System Reaches 92% MBPP via LLM-Symbolic Engine Routing

Chen, Ho Yiing · 2026-06-05 · Zenodo

doi:10.5281/zenodo.20557449 · PDF

Abstract

Bragi-LLM is a sub-gigabyte (805 MB total) Python code-generation system that reaches 92% pass rate on the MBPP test split under 100-problem sustained evaluation (single-shot, greedy decoding, no retry). It places within 2 absolute points of Qwen2.5-Coder-7B fp16 (94%, 14 GB, 17x the footprint), and 27 absolute points above the same 1.5B-Q3 backbone evaluated standalone (65%). The system combines (i) a 786 MB Q3_K_M quantised Qwen2.5-Coder-1.5B-Instruct backbone with imatrix calibration, (ii) a 15 KB hand-engineered symbolic engine library of 50 unit-tested helpers covering rare formulas, and

Chen, Ho Yiing (norika) · Independent Researcher, Taiwan · ORCID 0009-0006-6816-9891