6
Models Tested
33.3%
Success Rate
58s
Avg Duration
28s - 1m 26s
Duration Range
| Score | Model | Duration | Session (KB) | test_calculator_functions.sh | test_file_exists.sh | test_main_output.sh |
|---|---|---|---|---|---|---|
| 100.0% | litellm/GLM-4.5-Air-FP8-dev | 1m 26s | 67.9 | ✅ | ✅ | ✅ |
| 100.0% | qwen/qwen3-coder | 1m 18s | 67.9 | ✅ | ✅ | ✅ |
| 0.0% | deepseek/deepseek-v3.1-terminus | 28s | 24.9 | ❌ | ❌ | ❌ |
| 0.0% | openai/gpt-oss-20b | 1m 3s | 108.3 | ❌ | ❌ | ❌ |
| 0.0% | openai/gpt-oss-120b | 1m 1s | 156.7 | ❌ | ❌ | ❌ |
| 0.0% | qwen/qwen3-14b | 30s | 24.9 | ❌ | ❌ | ❌ |