CodePanel

Ruby code review: which models catch the bugs

Of the 17 models ranked on Ruby code, 5 models tied for the most known bugs: 100%. 9 of the other 12 models have 95% intervals that overlap theirs. Gemini 3.5 Flash-Lite caught the fewest, 0%. Together they caught 69% of known bugs in Ruby code, against 60% for the same 17 across the whole benchmark.

Updated 2026-11-04Full leaderboardMethodology

Results

Known defects caught, with 95% intervals. The tick is the model's recall over every bug.

Ruby code review: which models catch the bugs
ModelCaught hereEvery bug
GPT-6.1 Sol100%89%
Gemini 4 Argon100%82%
Claude Sonnet 5.5100%77%
Gemini 3.8 Flash100%68%
Gemini 3.1 Pro100%67%
Claude Opus 5.589%[67%–100%]90%
GPT-5.6 Sol67%[0%–100%]72%
GPT-6 Luna67%[0%–100%]67%
Claude Haiku 5.567%[0%–100%]64%
GLM-5.3 Flash (Workers AI)67%[33%–100%]61%
GLM-5.3 (Workers AI)67%[0%–100%]61%
DeepSeek V4 Pro (Workers AI)67%[50%–100%]54%
Gemma 4 26B A4B (Workers AI)67%[0%–100%]26%
DeepSeek V4 Flash (Workers AI)44%[0%–67%]49%
GPT-OSS 120B (Workers AI)33%[0%–67%]40%
Kimi K2.7 Code (Workers AI)33%[0%–100%]38%
Gemini 3.5 Flash-Lite0%11%

Against the rest of the benchmark

Ruby was the fifth-easiest of the 12 languages for these models.

Gemma 4 26B A4B gains the most here: 67%, against 26% over every bug.

Gemini 3.5 Flash-Lite drops the most: 0%, against 11% over every bug.

A test PR that split the models

The change “Reimplement Action Cable redis adapter with redis-client” in rails/rails, a change of 100 to 300 lines that landed in June 2026, introduced a logic bug that a later upstream commit fixed.

GPT-6.1 Sol, Gemini 4 Argon, Claude Sonnet 5.5, Gemini 3.8 Flash, GPT-6 Luna, Gemini 3.1 Pro and Gemma 4 26B A4B caught it; Claude Opus 5.5, GLM-5.3 Flash and DeepSeek V4 Pro (Workers AI) caught it in part; GPT-5.6 Sol, Claude Haiku 5.5, GLM-5.3, DeepSeek V4 Flash (Workers AI), GPT-OSS 120B (Workers AI), Kimi K2.7 Code (Workers AI) and Gemini 3.5 Flash-Lite missed it.