OpenCodeInterpreter-DS-6.7B: benchmark results
Provider: Other.
Aggregate Rating 1440 ± 43, rank #4206 of 10916 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| CIBER (OpenCodeInterpreter) | 49.3 | Attack success rate (%) of memory-poisoning, indirect-inject | 100 |
| CIBER (OpenCodeInterpreter) - Direct Prompt Injection | 50.4 | Attack success rate (%) of direct prompt-injection attacks a | 100 |
| CIBER (OpenCodeInterpreter) - Prompt-Based Backdoor | 0 | Attack success rate (%) of prompt-based backdoor attacks aga | 100 |
| BigCode Models Leaderboard | 73.2 | HumanEval Python Pass@1 (%) | 86.4 |
| EvalPlus (HumanEval+ & MBPP+) | 69.2 | Pass@1 avg (%) | 79.8 |
| CodeEditorBench Plus - Code Translate (Three-shot) | 37.2 | pass@1 (%, greedy decoding, three-shot, CodeEditorBench_Plus | 77.8 |
| CIBER (OpenCodeInterpreter) - Memory Poisoning | 67.5 | Attack success rate (%) of memory-poisoning attacks against | 66.7 |
| CodeEditorBench Plus - Code Requirement Switch (Three-shot) | 16.5 | pass@1 (%, greedy decoding, three-shot, CodeEditorBench_Plus | 66.7 |
| CodeEditorBench Plus - Code Requirement Switch (Zero-shot) | 12.6 | pass@1 (%, greedy decoding, zero-shot, CodeEditorBench_Plus) | 66.7 |
| McEval | 46 | AVG (%) | 64.3 |
| CodeEditorBench Plus - Code Translate (Zero-shot) | 35.7 | pass@1 (%, greedy decoding, zero-shot, CodeEditorBench_Plus) | 61.1 |
| EffiBench - NMU | 1.91 | Normalized Memory Usage | 58.5 |
Interactive version: theaggregate.ai/model?slug=opencodeinterpreter-ds-6-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.