{"benchmark_id":"automationbench","benchmark_name":"AutomationBench","benchmark_description":"AutomationBench is a tool-use benchmark that evaluates AI agents on automating real-world workflows, testing their ability to orchestrate tools and complete multi-step automation tasks.","max_score":1.0,"categories":["reasoning","agents","tool_calling"],"modality":"text","total_models":23,"entries":[{"rank":1,"model_id":"mistral-large-4","model_name":"Mistral Large 4","organization_name":"Mistral AI","organization_id":"mistral","benchmark_score":0.599,"normalized_score":0.599,"verified":false,"self_reported":true,"provider_id":"mistral","input_cost_per_million":0.68,"output_cost_per_million":2.09,"speed_rps":null,"context_window":1000000,"release_date":"2026-10-06","announcement_date":"2026-10-06","multimodal":true,"param_count":1050000000000,"is_new":false},{"rank":2,"model_id":"deepseek-v4.1-flash","model_name":"DeepSeek-V4.1-Flash","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.5479999999999999,"normalized_score":0.5479999999999999,"verified":false,"self_reported":true,"provider_id":"fireworks","input_cost_per_million":0.22,"output_cost_per_million":0.66,"speed_rps":null,"context_window":1040000,"release_date":"2026-09-10","announcement_date":"2026-09-10","multimodal":true,"param_count":763205315794,"is_new":false},{"rank":3,"model_id":"gemini-4-argon","model_name":"Gemini 4 Argon","organization_name":"Google","organization_id":"google","benchmark_score":0.513,"normalized_score":0.513,"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":"2026-09-30","announcement_date":"2026-09-30","multimodal":true,"param_count":null,"is_new":false},{"rank":4,"model_id":"muse-spark-1.3","model_name":"Muse Spark 1.3","organization_name":"Meta","organization_id":"meta","benchmark_score":0.494,"normalized_score":0.494,"verified":false,"self_reported":true,"provider_id":"meta","input_cost_per_million":1.25,"output_cost_per_million":4.25,"speed_rps":null,"context_window":1048576,"release_date":"2026-09-02","announcement_date":"2026-09-02","multimodal":true,"param_count":null,"is_new":false},{"rank":5,"model_id":"glm-5.3-flash","model_name":"GLM-5.3-Flash","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.488,"normalized_score":0.488,"verified":false,"self_reported":true,"provider_id":"z","input_cost_per_million":0.15,"output_cost_per_million":0.5,"speed_rps":null,"context_window":1000000,"release_date":"2026-08-26","announcement_date":"2026-08-26","multimodal":true,"param_count":320000000000,"is_new":false},{"rank":6,"model_id":"glm-5.3","model_name":"GLM-5.3","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.482,"normalized_score":0.482,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":1.2,"output_cost_per_million":4.0,"speed_rps":null,"context_window":1048576,"release_date":"2026-08-14","announcement_date":"2026-08-14","multimodal":false,"param_count":753000000000,"is_new":false},{"rank":7,"model_id":"claude-sonnet-5-5","model_name":"Claude Sonnet 5.5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.447,"normalized_score":0.447,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":2.0,"output_cost_per_million":10.0,"speed_rps":42.0,"context_window":1000000,"release_date":"2026-09-28","announcement_date":"2026-09-28","multimodal":true,"param_count":null,"is_new":false},{"rank":8,"model_id":"gpt-6-astra","model_name":"GPT-6 Astra","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.414,"normalized_score":0.414,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":10.0,"output_cost_per_million":50.0,"speed_rps":null,"context_window":1050000,"release_date":"2026-09-04","announcement_date":"2026-09-03","multimodal":true,"param_count":null,"is_new":false},{"rank":9,"model_id":"claude-opus-5-5","model_name":"Claude Opus 5.5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.4,"normalized_score":0.4,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":4.0,"output_cost_per_million":20.0,"speed_rps":null,"context_window":1000000,"release_date":"2026-09-22","announcement_date":"2026-09-22","multimodal":true,"param_count":null,"is_new":false},{"rank":10,"model_id":"hy4-preview","model_name":"Hy4 preview","organization_name":"Tencent","organization_id":"tencent","benchmark_score":0.321,"normalized_score":0.321,"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":"2026-08-28","announcement_date":"2026-08-28","multimodal":false,"param_count":770000000000,"is_new":false},{"rank":11,"model_id":"deepseek-v4-pro-0813","model_name":"DeepSeek-V4-Pro-0813","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.318,"normalized_score":0.318,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":1.3,"output_cost_per_million":2.6,"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":12,"model_id":"claude-fable-5-1","model_name":"Claude Fable 5.1","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.314,"normalized_score":0.314,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":10.0,"output_cost_per_million":50.0,"speed_rps":null,"context_window":1000000,"release_date":"2026-09-01","announcement_date":"2026-09-01","multimodal":true,"param_count":null,"is_new":false},{"rank":13,"model_id":"kimi-k3","model_name":"Kimi K3","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.308,"normalized_score":0.308,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":2.85,"output_cost_per_million":14.25,"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":14,"model_id":"gemini-3.7-flash","model_name":"Gemini 3.7 Flash","organization_name":"Google","organization_id":"google","benchmark_score":0.304,"normalized_score":0.304,"verified":false,"self_reported":true,"provider_id":"google","input_cost_per_million":0.75,"output_cost_per_million":3.75,"speed_rps":null,"context_window":1048576,"release_date":"2026-08-13","announcement_date":"2026-08-13","multimodal":true,"param_count":null,"is_new":false},{"rank":15,"model_id":"qwen3.8-max","model_name":"Qwen3.8 Max","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.273,"normalized_score":0.273,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":1.65,"output_cost_per_million":4.951,"speed_rps":null,"context_window":256000,"release_date":"2026-08-02","announcement_date":"2026-08-02","multimodal":true,"param_count":2400000000000,"is_new":false},{"rank":16,"model_id":"claude-opus-5","model_name":"Claude Opus 5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.26,"normalized_score":0.26,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":5.0,"output_cost_per_million":25.0,"speed_rps":null,"context_window":1000000,"release_date":"2026-07-24","announcement_date":"2026-07-24","multimodal":true,"param_count":null,"is_new":false},{"rank":17,"model_id":"deepseek-v4-flash-vision-exp","model_name":"DeepSeek-V4-Flash-Vision-Exp","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.257,"normalized_score":0.257,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.44,"output_cost_per_million":1.32,"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":18,"model_id":"deepseek-v4-flash-0731","model_name":"DeepSeek-V4-Flash-0731","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.251,"normalized_score":0.251,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.06,"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":19,"model_id":"gpt-5.6-sol","model_name":"GPT-5.6 Sol","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.181,"normalized_score":0.181,"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":20,"model_id":"claude-fable-5","model_name":"Claude Fable 5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.174,"normalized_score":0.174,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":10.0,"output_cost_per_million":50.0,"speed_rps":null,"context_window":1000000,"release_date":"2026-06-09","announcement_date":"2026-06-09","multimodal":true,"param_count":null,"is_new":false},{"rank":21,"model_id":"gpt-5.6-terra","model_name":"GPT-5.6 Terra","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.152,"normalized_score":0.152,"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":22,"model_id":"gpt-5.6-luna","model_name":"GPT-5.6 Luna","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.149,"normalized_score":0.149,"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":23,"model_id":"claude-sonnet-5","model_name":"Claude Sonnet 5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.135,"normalized_score":0.135,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":2.0,"output_cost_per_million":10.0,"speed_rps":42.0,"context_window":1000000,"release_date":"2026-06-30","announcement_date":"2026-06-30","multimodal":true,"param_count":null,"is_new":false}]}