{"benchmark_id":"deepswe","benchmark_name":"DeepSWE","benchmark_description":"DeepSWE is a software engineering agent benchmark evaluated with the mini-swe-agent harness, where each task is solved in an isolated container with no internet access. It measures an agent's ability to autonomously resolve real-world coding issues end to end.","max_score":1.0,"categories":["agents","code"],"modality":"text","total_models":12,"entries":[{"rank":1,"model_id":"gpt-5.6-sol","model_name":"GPT-5.6 Sol","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.727,"normalized_score":0.727,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":5.0,"output_cost_per_million":30.0,"speed_rps":null,"context_window":1050000,"release_date":"2026-07-09","announcement_date":"2026-07-09","multimodal":true,"param_count":null,"is_new":false},{"rank":2,"model_id":"gpt-5.6-terra","model_name":"GPT-5.6 Terra","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.696,"normalized_score":0.696,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":2.0,"output_cost_per_million":12.0,"speed_rps":null,"context_window":1050000,"release_date":"2026-07-09","announcement_date":"2026-07-09","multimodal":true,"param_count":null,"is_new":false},{"rank":3,"model_id":"kimi-k3","model_name":"Kimi K3","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.675,"normalized_score":0.675,"verified":false,"self_reported":true,"provider_id":"novita","input_cost_per_million":3.0,"output_cost_per_million":15.0,"speed_rps":null,"context_window":1048576,"release_date":"2026-07-16","announcement_date":"2026-07-16","multimodal":true,"param_count":2800000000000,"is_new":false},{"rank":4,"model_id":"gpt-5.6-luna","model_name":"GPT-5.6 Luna","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.672,"normalized_score":0.672,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":0.2,"output_cost_per_million":1.2,"speed_rps":null,"context_window":1050000,"release_date":"2026-07-09","announcement_date":"2026-07-09","multimodal":true,"param_count":null,"is_new":false},{"rank":5,"model_id":"deepseek-v4-pro-0813","model_name":"DeepSeek-V4-Pro-0813","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.627,"normalized_score":0.627,"verified":false,"self_reported":true,"provider_id":"deepseek","input_cost_per_million":0.435,"output_cost_per_million":0.87,"speed_rps":null,"context_window":1048576,"release_date":"2026-08-13","announcement_date":"2026-08-13","multimodal":false,"param_count":1600000000000,"is_new":false},{"rank":6,"model_id":"deepseek-v4-flash-vision-exp","model_name":"DeepSeek-V4-Flash-Vision-Exp","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.593,"normalized_score":0.593,"verified":false,"self_reported":true,"provider_id":"deepseek","input_cost_per_million":0.22,"output_cost_per_million":0.66,"speed_rps":null,"context_window":1048576,"release_date":"2026-08-21","announcement_date":"2026-08-21","multimodal":true,"param_count":null,"is_new":false},{"rank":7,"model_id":"deepseek-v4-flash-0731","model_name":"DeepSeek-V4-Flash-0731","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.544,"normalized_score":0.544,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.09,"output_cost_per_million":0.18,"speed_rps":null,"context_window":1048576,"release_date":"2026-07-31","announcement_date":"2026-07-31","multimodal":false,"param_count":304000000000,"is_new":false},{"rank":8,"model_id":"grok-4.5","model_name":"Grok 4.5","organization_name":"xAI","organization_id":"xai","benchmark_score":0.53,"normalized_score":0.53,"verified":false,"self_reported":true,"provider_id":"xai","input_cost_per_million":2.0,"output_cost_per_million":6.0,"speed_rps":80.0,"context_window":500000,"release_date":"2026-07-16","announcement_date":"2026-07-16","multimodal":true,"param_count":null,"is_new":false},{"rank":9,"model_id":"glm-5.2","model_name":"GLM-5.2","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.462,"normalized_score":0.462,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.95,"output_cost_per_million":3.0,"speed_rps":null,"context_window":1048576,"release_date":"2026-06-16","announcement_date":"2026-06-16","multimodal":false,"param_count":753000000000,"is_new":false},{"rank":10,"model_id":"seed-2.1-pro","model_name":"Seed 2.1 Pro","organization_name":"ByteDance","organization_id":"bytedance","benchmark_score":0.327,"normalized_score":0.327,"verified":false,"self_reported":true,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"speed_rps":null,"context_window":null,"release_date":"2026-06-24","announcement_date":"2026-06-24","multimodal":true,"param_count":null,"is_new":false},{"rank":11,"model_id":"hy3","model_name":"Hy3","organization_name":"Tencent","organization_id":"tencent","benchmark_score":0.28,"normalized_score":0.28,"verified":false,"self_reported":true,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"speed_rps":null,"context_window":null,"release_date":"2026-07-06","announcement_date":"2026-07-06","multimodal":false,"param_count":295000000000,"is_new":false},{"rank":12,"model_id":"seed-2.1-turbo","model_name":"Seed 2.1 Turbo","organization_name":"ByteDance","organization_id":"bytedance","benchmark_score":0.23,"normalized_score":0.23,"verified":false,"self_reported":true,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"speed_rps":null,"context_window":null,"release_date":"2026-06-24","announcement_date":"2026-06-24","multimodal":true,"param_count":null,"is_new":false}]}