Run of 2025-12-24 15:00:21

24
Models Tested
13
Total Tasks
47.1%
Overall Success Rate
147/312
Passed Combinations

Details

Score Model task10_multiple_tests task11_relationship_classifier task12_need_reply task13_meeting_action_items task14_graph_money_distribution task1_file_list task2_fix_python_syntax task4_web_fetch task5_dedup_contact task6_config_merger task7_log_parser task8_regex_extraction task9_cpp_footguns
84.6% openrouter/google/gemini-3-pro-preview
84.6% openrouter/anthropic/claude-opus-4.5
76.9% openrouter/anthropic/claude-haiku-4.5
76.9% openrouter/openai/gpt-5.2
76.9% openrouter/openai/gpt-5-mini
69.2% openrouter/google/gemini-2.5-pro
69.2% litellm/GLM-4.5-Air-FP8-dev
69.2% openrouter/anthropic/claude-sonnet-4.5
61.5% openrouter/openai/gpt-5
61.5% openrouter/qwen/qwen3-coder
61.5% openrouter/deepseek/deepseek-v3.1-terminus
61.5% openrouter/x-ai/grok-code-fast-1
46.2% openrouter/openai/gpt-4.1-mini
38.5% openrouter/openai/gpt-5-nano
38.5% openrouter/x-ai/grok-3-mini
38.5% openrouter/google/gemini-2.5-flash-lite-preview-09-2025
23.1% openrouter/google/gemini-2.5-flash-preview-09-2025
23.1% openrouter/openai/gpt-oss-20b
23.1% openrouter/deepseek/deepseek-chat-v3-0324
15.4% openrouter/openai/gpt-oss-120b
15.4% openrouter/openai/gpt-4o-mini
15.4% openrouter/openai/gpt-4.1-nano
0.0% litellm/DeepSeek-V3.2-sandbox
0.0% litellm/GLM-4.6-trtllm-sandbox