CodeGemma
Google DeepMind — Gemma
rich detail 26 method fields published
Evaluation
HumanEval Infilling
single-line and multi-line metrics in the HumanEval Infilling benchmarks introduced in Fried et al. (2023); latency measured as total seconds to obtain 128-token continuations per task
How it was run
Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.