LLM Evaluation

← Run of 2025-10-24 15:00:17
24
Models Tested
6
Total Tasks
54.2%
Overall Success Rate
78/144
Passed Combinations
Score Model task1_file_list task2_fix_python_syntax task4_web_fetch task7_log_parser task8_regex_extraction test_multiple_tests
100.0% openrouter/anthropic/claude-sonnet-4.5 ✅ ✅ ✅ ✅ ✅ ✅
100.0% openrouter/anthropic/claude-sonnet-4 ✅ ✅ ✅ ✅ ✅ ✅
83.3% openrouter/openai/gpt-5 ✅ ✅ ✅ ❌ ✅ ✅
83.3% openrouter/anthropic/claude-3.5-sonnet ✅ ✅ ❌ ✅ ✅ ✅
83.3% openrouter/anthropic/claude-3.7-sonnet ✅ ✅ ❌ ✅ ✅ ✅
83.3% openrouter/anthropic/claude-haiku-4.5 ✅ ✅ ❌ ✅ ✅ ✅
83.3% openrouter/openai/gpt-5-mini ✅ ✅ ❌ ✅ ✅ ✅
83.3% openrouter/openai/gpt-4.1-mini ✅ ✅ ✅ ❌ ✅ ✅
66.7% openrouter/openai/gpt-5-nano ✅ ✅ ❌ ❌ ✅ ✅
66.7% openrouter/anthropic/claude-3.5-haiku ✅ ✅ ❌ ❌ ✅ ✅
50.0% openrouter/google/gemini-2.5-flash-preview-09-2025 ✅ ❌ ❌ ✅ ❌ ✅
50.0% openrouter/qwen/qwen3-coder ✅ ✅ ❌ ❌ ❌ ✅
50.0% openrouter/x-ai/grok-3-mini ❌ ✅ ❌ ✅ ❌ ✅
50.0% openrouter/openai/gpt-4o-mini ✅ ❌ ❌ ❌ ✅ ✅
50.0% openrouter/openai/gpt-oss-20b ✅ ❌ ❌ ❌ ✅ ✅
50.0% openrouter/deepseek/deepseek-v3.1-terminus ✅ ❌ ❌ ❌ ✅ ✅
50.0% litellm/GLM-4.5-Air-FP8-dev ❌ ✅ ❌ ❌ ✅ ✅
33.3% openrouter/anthropic/claude-3-haiku ✅ ❌ ❌ ✅ ❌ ❌
33.3% openrouter/google/gemini-2.5-pro ✅ ❌ ❌ ✅ ❌ ❌
33.3% openrouter/openai/gpt-4.1-nano ✅ ❌ ❌ ❌ ❌ ✅
16.7% openrouter/google/gemini-2.5-flash-lite-preview-09-2025 ❌ ❌ ❌ ❌ ❌ ✅
0.0% openrouter/openai/gpt-oss-120b ❌ ❌ ❌ ❌ ❌ ❌
0.0% openrouter/deepseek/deepseek-chat-v3-0324 ❌ ❌ ❌ ❌ ❌ ❌
0.0% openrouter/qwen/qwen3-14b ❌ ❌ ❌ ❌ ❌ ❌