Everyone-Coder-33B-Base: benchmark results
A coding model merge assembled by rombodawg from DeepSeek-Coder-33B-Instruct, CodeFuse-DeepSeek-33B and WizardCoder-33B-V1.1. Provider: rombodawg. Access: Open.
Aggregate Rating 1271 ± 36, rank #9050 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 39.8 | Accuracy (%) (5-shot) | 54.9 |
| Open LLM Leaderboard v1 - MMLU | 44.05 | Accuracy (%) (5-shot) | 23.1 |
| Open LLM Leaderboard v1 - ARC Challenge | 45.99 | Normalized accuracy (%) (25-shot) | 21.2 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 42.26 | MC2 (%) (0-shot) | 19.9 |
| Open LLM Leaderboard v1 - WinoGrande | 63.06 | Accuracy (%) (5-shot) | 15.9 |
| Open LLM Leaderboard v1 - HellaSwag | 61.71 | Normalized accuracy (%) (10-shot) | 15.6 |
Interactive version: theaggregate.ai/model?slug=everyone-coder-33b-base · How It Works · Data refreshed daily, snapshot 2026-11-08.