CodeGen-16B-Multi: benchmark results
Provider: Salesforce. Released 2022-04-13. Access: Open.
Unified ELO 1156 ± 43, rank #2645 of 2656 rated models, from 13 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ToolBench - Google Sheets | 21.4 | Task Score | 53.5 |
| ToolBench - WebShop Short | 0.5 | Task Score | 51.2 |
| ToolBench - Home Search | 47 | Task Score | 46.5 |
| ToolBench - The Cat API | 75 | Task Score | 45.3 |
| ToolBench - WebShop Long | 0 | Task Score | 44.2 |
| ToolBench Leaderboard | 28.76 | Average Task Score | 44.2 |
| ToolBench - Tabletop | 8.6 | Task Score | 40.7 |
| ToolBench - Open Weather | 56 | Task Score | 39.5 |
| ToolBench - VirtualHome | 14.1 | Task Score | 37.2 |
| ToolBench - Trip Booking | 7.5 | Task Score | 31.4 |
| SAFIM | 30.99 | Average | 28.6 |
| BigCode Models Leaderboard | 19.3 | HumanEval Python Pass@1 (%) | 4.2 |
Interactive version: theaggregate.ai/model?slug=codegen-16b-multi · How It Works · Data refreshed daily, snapshot 2026-09-19.