{"benchmark_id":"zerobench","name":"ZEROBench","parent_benchmark":null,"categories":["multimodal","reasoning","vision"],"modality":"image","multilingual":false,"max_score":100.0,"language":"en","description":"ZEROBench is a challenging vision benchmark designed to test models on zero-shot visual understanding tasks.","paper_link":null,"implementation_link":null,"verified":false,"created_at":"2026-05-07T16:53:27.058875+00:00","updated_at":"2026-10-31T18:28:42.893517+00:00","statistics":{"total_models":16,"average_score":0.9225,"min_score":0.04,"max_score":11.0,"score_stddev":2.692870834877405,"verified_count":0,"self_reported_count":16},"child_benchmarks":[{"benchmark_id":"zerobench-sub","name":"ZEROBench-Sub","categories":["multimodal","reasoning","vision"],"modality":"image","max_score":100.0,"description":"ZEROBench-Sub is a subset of the ZEROBench benchmark."}],"linked_dataset":null,"models":[{"rank":1,"model_id":"seed-1.8","model_name":"Seed 1.8","organization_id":"bytedance","organization_name":"ByteDance","organization_country":"CN","score":11.0,"normalized_score":0.11,"verified":false,"self_reported":true,"self_reported_source":"https://lf3-static.bytednsdoc.com/obj/eden-cn/lapzild-tss/ljhwZthlaukjlkulzlp/research/Seed-1.8-Modelcard.pdf","analysis_method":"Official Table 2; think-high; Pass@1","verification_date":null,"provider_id":"deepinfra","input_cost_per_million":0.25,"output_cost_per_million":2.0,"context_window":256000,"announcement_date":"2026-02-17","param_count":null,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":"DeepInfra","best_throughput":null,"throughput_provider":"DeepInfra","context_provider":"DeepInfra"},{"rank":2,"model_id":"deepseek-v4.1-flash-1017","model_name":"DeepSeek-V4.1-Flash-1017","organization_id":"deepseek","organization_name":"DeepSeek","organization_country":"CN","score":0.51,"normalized_score":0.51,"verified":false,"self_reported":true,"self_reported_source":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017","analysis_method":"Pass@5; ZeroBench-main with tools. Self-reported on Hugging Face model card and DeepSeek API changelog 2026-10-17.","verification_date":null,"provider_id":"deepseek","input_cost_per_million":0.3,"output_cost_per_million":1.2,"context_window":1048576,"announcement_date":"2026-10-17","param_count":763205315794,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":"DeepSeek","best_throughput":null,"throughput_provider":"DeepSeek","context_provider":"DeepSeek"},{"rank":3,"model_id":"muse-spark-open","model_name":"Muse Spark Open","organization_id":"meta","organization_name":"Meta","organization_country":"US","score":0.5,"normalized_score":0.5,"verified":false,"self_reported":true,"self_reported_source":"https://huggingface.co/meta-models/Muse-Spark-Open-816B-A42B","analysis_method":"HF Muse Spark Open model card; xhigh. ZeroBench (w/ tools) 50.0.","verification_date":null,"provider_id":"together","input_cost_per_million":1.25,"output_cost_per_million":4.25,"context_window":1048576,"announcement_date":"2026-10-15","param_count":816000000000,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":"Together","best_throughput":null,"throughput_provider":"Together","context_provider":"Together"},{"rank":4,"model_id":"deepseek-v4.1-flash","model_name":"DeepSeek-V4.1-Flash","organization_id":"deepseek","organization_name":"DeepSeek","organization_country":"CN","score":0.49,"normalized_score":0.49,"verified":false,"self_reported":true,"self_reported_source":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash","analysis_method":"Pass@5; main split with tools; Claude Code harness; 512K-token context. Maximum reasoning effort (reasoning_effort=100); temperature=1.0; top_p=0.95.","verification_date":null,"provider_id":"fireworks","input_cost_per_million":0.22,"output_cost_per_million":0.66,"context_window":1048576,"announcement_date":"2026-09-10","param_count":763205315794,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":"Fireworks","best_throughput":null,"throughput_provider":"Fireworks","context_provider":"Novita"},{"rank":5,"model_id":"kimi-k3","model_name":"Kimi K3","organization_id":"moonshotai","organization_name":"Moonshot AI","organization_country":"CN","score":0.41,"normalized_score":0.41,"verified":false,"self_reported":true,"self_reported_source":"https://www.kimi.com/blog/kimi-k3","analysis_method":"Main split with Python; pass@5; 5 runs","verification_date":null,"provider_id":"deepinfra","input_cost_per_million":2.85,"output_cost_per_million":14.25,"context_window":1048576,"announcement_date":"2026-07-16","param_count":2800000000000,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":"DeepInfra","best_throughput":null,"throughput_provider":"DeepInfra","context_provider":"Novita"},{"rank":6,"model_id":"kimi-k3.1-pro","model_name":"Kimi K3.1 Pro","organization_id":"moonshotai","organization_name":"Moonshot AI","organization_country":"CN","score":0.39,"normalized_score":0.39,"verified":false,"self_reported":true,"self_reported_source":"https://www.kimi.ai/blog/kimi-k3-1-pro","analysis_method":"Main split; pass@5; 5 runs; Max reasoning effort; temperature 1.0, top-p 0.95","verification_date":null,"provider_id":"moonshot","input_cost_per_million":2.0,"output_cost_per_million":10.0,"context_window":1048576,"announcement_date":"2026-10-23","param_count":null,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":"Moonshot AI","best_throughput":null,"throughput_provider":"Moonshot AI","context_provider":"Moonshot AI"},{"rank":7,"model_id":"deepseek-v4-flash-vision-exp","model_name":"DeepSeek-V4-Flash-Vision-Exp","organization_id":"deepseek","organization_name":"DeepSeek","organization_country":"CN","score":0.35,"normalized_score":0.35,"verified":false,"self_reported":true,"self_reported_source":"https://api-docs.deepseek.com/updates/","analysis_method":"Pass@5; self-reported on DeepSeek API changelog 2026-08-21.","verification_date":null,"provider_id":"deepinfra","input_cost_per_million":0.44,"output_cost_per_million":1.32,"context_window":1048576,"announcement_date":"2026-08-21","param_count":null,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":"DeepInfra","best_throughput":null,"throughput_provider":"DeepInfra","context_provider":"DeepInfra"},{"rank":8,"model_id":"muse-spark","model_name":"Muse Spark","organization_id":"meta","organization_name":"Meta","organization_country":"US","score":0.33,"normalized_score":0.33,"verified":false,"self_reported":true,"self_reported_source":"https://ai.meta.com/static-resource/muse-spark-eval-methodology","analysis_method":"Thinking mode, with Python, pass@5","verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2026-04-08","param_count":null,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":9,"model_id":"seed-2.1-pro","model_name":"Seed 2.1 Pro","organization_id":"bytedance","organization_name":"ByteDance","organization_country":"CN","score":0.18,"normalized_score":0.18,"verified":false,"self_reported":true,"self_reported_source":"https://seed.bytedance.com/en/seed2_1","analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2026-06-24","param_count":null,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":10,"model_id":"seed-2.1-turbo","model_name":"Seed 2.1 Turbo","organization_id":"bytedance","organization_name":"ByteDance","organization_country":"CN","score":0.11,"normalized_score":0.11,"verified":false,"self_reported":true,"self_reported_source":"https://seed.bytedance.com/en/seed2_1","analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2026-06-24","param_count":null,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":10,"model_id":"kimi-k2.5","model_name":"Kimi K2.5","organization_id":"moonshotai","organization_name":"Moonshot AI","organization_country":"CN","score":0.11,"normalized_score":0.11,"verified":false,"self_reported":true,"self_reported_source":"https://www.kimi.com/blog/kimi-k2-5.html","analysis_method":"Image","verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2026-01-27","param_count":1000000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":12,"model_id":"qwen3.5-27b","model_name":"Qwen3.5-27B","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.1,"normalized_score":0.1,"verified":false,"self_reported":true,"self_reported_source":"https://qwen.ai/blog?id=qwen3.5","analysis_method":null,"verification_date":null,"provider_id":"deepinfra","input_cost_per_million":0.26,"output_cost_per_million":2.6,"context_window":262144,"announcement_date":"2026-02-24","param_count":27000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":"DeepInfra","best_throughput":null,"throughput_provider":"DeepInfra","context_provider":"Novita"},{"rank":13,"model_id":"qwen3.5-122b-a10b","model_name":"Qwen3.5-122B-A10B","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.09,"normalized_score":0.09,"verified":false,"self_reported":true,"self_reported_source":"https://qwen.ai/blog?id=qwen3.5","analysis_method":null,"verification_date":null,"provider_id":"deepinfra","input_cost_per_million":0.29,"output_cost_per_million":2.4,"context_window":262144,"announcement_date":"2026-02-24","param_count":122000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":"DeepInfra","best_throughput":null,"throughput_provider":"DeepInfra","context_provider":"DeepInfra"},{"rank":14,"model_id":"qwen3.5-35b-a3b","model_name":"Qwen3.5-35B-A3B","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.08,"normalized_score":0.08,"verified":false,"self_reported":true,"self_reported_source":"https://qwen.ai/blog?id=qwen3.5","analysis_method":null,"verification_date":null,"provider_id":"deepinfra","input_cost_per_million":0.14,"output_cost_per_million":1.0,"context_window":262144,"announcement_date":"2026-02-24","param_count":35000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":"DeepInfra","best_throughput":null,"throughput_provider":"DeepInfra","context_provider":"DeepInfra"},{"rank":15,"model_id":"seed-2.0-mini","model_name":"Seed 2.0 Mini","organization_id":"bytedance","organization_name":"ByteDance","organization_country":"CN","score":0.07,"normalized_score":0.07,"verified":false,"self_reported":true,"self_reported_source":"https://lf3-static.bytednsdoc.com/obj/eden-cn/lapzild-tss/ljhwZthlaukjlkulzlp/seed2/0214/Seed2.0%20Model%20Card.pdf","analysis_method":"Model Card Table 8; ZEROBench main split percentage score.","verification_date":null,"provider_id":"deepinfra","input_cost_per_million":0.1,"output_cost_per_million":0.4,"context_window":256000,"announcement_date":"2026-02-17","param_count":null,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":"DeepInfra","best_throughput":null,"throughput_provider":"DeepInfra","context_provider":"DeepInfra"},{"rank":16,"model_id":"qwen3-vl-235b-a22b-thinking","model_name":"Qwen3 VL 235B A22B Thinking","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.04,"normalized_score":0.04,"verified":false,"self_reported":true,"self_reported_source":null,"analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2025-09-22","param_count":236000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null}]}