6
Models Tested
50.0%
Success Rate
53s
Avg Duration
40s - 1m 2s
Duration Range
| Score | Model | Duration | Session (KB) | test_1_file_exists.sh | test_2_valid_json.sh | test_3_json_structure.sh | test_4_expected_content.sh |
|---|---|---|---|---|---|---|---|
| 100.0% | openai/gpt-oss-20b | 1m 2s | 420.3 | ✅ | ✅ | ✅ | ✅ |
| 100.0% | litellm/GLM-4.5-Air-FP8-dev | 1m 2s | 91.8 | ✅ | ✅ | ✅ | ✅ |
| 100.0% | qwen/qwen3-coder | 59s | 64.7 | ✅ | ✅ | ✅ | ✅ |
| 0.0% | deepseek/deepseek-v3.1-terminus | 41s | 25.0 | ❌ | ❌ | ❌ | ❌ |
| 0.0% | openai/gpt-oss-120b | 55s | 175.6 | ❌ | ❌ | ❌ | ❌ |
| 0.0% | qwen/qwen3-14b | 40s | 24.9 | ❌ | ❌ | ❌ | ❌ |