{"benchmark_id":"seccodebench","benchmark_name":"SecCodeBench","benchmark_description":"SecCodeBench evaluates LLM coding agents on secure code generation and vulnerability detection, testing the ability to produce code that is both functional and free from security vulnerabilities.","max_score":1.0,"categories":["code"],"modality":"text","total_models":2,"entries":[{"rank":1,"model_id":"qwen3.5-397b-a17b","model_name":"Qwen3.5-397B-A17B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.683,"normalized_score":0.683,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.45,"output_cost_per_million":3.0,"speed_rps":null,"context_window":262144,"release_date":"2026-02-16","announcement_date":"2026-02-16","multimodal":true,"param_count":397000000000,"is_new":false},{"rank":2,"model_id":"qwen3-max-thinking","model_name":"Qwen3 Max Thinking","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.575,"normalized_score":0.575,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":1.2,"output_cost_per_million":6.0,"speed_rps":null,"context_window":256000,"release_date":"2026-02-13","announcement_date":"2026-02-13","multimodal":false,"param_count":1000000000000,"is_new":false}]}