{"benchmark_id":"ifeval","benchmark_name":"IFEval","benchmark_description":"Instruction-Following Evaluation (IFEval) benchmark for large language models, focusing on verifiable instructions with 25 types of instructions and around 500 prompts containing one or more verifiable constraints","max_score":1.0,"categories":["chat","structured_output","instruction_following","general"],"modality":"text","total_models":68,"entries":[{"rank":1,"model_id":"qwen3.5-27b","model_name":"Qwen3.5-27B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.95,"normalized_score":0.95,"verified":false,"self_reported":true,"provider_id":"novita","input_cost_per_million":0.3,"output_cost_per_million":2.4,"speed_rps":null,"context_window":262144,"release_date":"2026-02-24","announcement_date":"2026-02-24","multimodal":true,"param_count":27000000000,"is_new":false},{"rank":2,"model_id":"qwen3.7-plus","model_name":"Qwen3.7-Plus","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.946,"normalized_score":0.946,"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":"2026-05-31","announcement_date":"2026-05-31","multimodal":true,"param_count":null,"is_new":false},{"rank":3,"model_id":"qwen3.7-max","model_name":"Qwen3.7 Max","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.943,"normalized_score":0.943,"verified":false,"self_reported":true,"provider_id":"novita","input_cost_per_million":1.25,"output_cost_per_million":3.75,"speed_rps":null,"context_window":1000000,"release_date":"2026-05-19","announcement_date":"2026-05-19","multimodal":false,"param_count":null,"is_new":false},{"rank":4,"model_id":"qwen3.6-plus","model_name":"Qwen3.6 Plus","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.943,"normalized_score":0.943,"verified":false,"self_reported":true,"provider_id":"together","input_cost_per_million":0.5,"output_cost_per_million":3.0,"speed_rps":null,"context_window":1000000,"release_date":"2026-03-31","announcement_date":"2026-04-02","multimodal":true,"param_count":null,"is_new":false},{"rank":5,"model_id":"o3-mini","model_name":"o3-mini","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.939,"normalized_score":0.939,"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":"2025-01-30","announcement_date":"2025-01-30","multimodal":false,"param_count":null,"is_new":false},{"rank":6,"model_id":"qwen3.5-122b-a10b","model_name":"Qwen3.5-122B-A10B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.934,"normalized_score":0.934,"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":"2026-02-24","announcement_date":"2026-02-24","multimodal":true,"param_count":122000000000,"is_new":false},{"rank":7,"model_id":"claude-3-7-sonnet-20250219","model_name":"Claude 3.7 Sonnet","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.932,"normalized_score":0.932,"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":"2025-02-24","announcement_date":"2025-02-24","multimodal":true,"param_count":null,"is_new":false},{"rank":8,"model_id":"qwen3.5-397b-a17b","model_name":"Qwen3.5-397B-A17B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.926,"normalized_score":0.926,"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":"2026-02-16","announcement_date":"2026-02-16","multimodal":true,"param_count":397000000000,"is_new":false},{"rank":9,"model_id":"llama-3.3-70b-instruct","model_name":"Llama 3.3 70B Instruct","organization_name":"Meta","organization_id":"meta","benchmark_score":0.921,"normalized_score":0.921,"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":"2024-12-06","announcement_date":"2024-12-06","multimodal":false,"param_count":70000000000,"is_new":false},{"rank":10,"model_id":"nova-pro","model_name":"Nova Pro","organization_name":"Amazon","organization_id":"amazon","benchmark_score":0.921,"normalized_score":0.921,"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":"2024-11-20","announcement_date":"2024-11-20","multimodal":true,"param_count":null,"is_new":false},{"rank":11,"model_id":"qwen3.5-35b-a3b","model_name":"Qwen3.5-35B-A3B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.919,"normalized_score":0.919,"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":"2026-02-24","announcement_date":"2026-02-24","multimodal":true,"param_count":35000000000,"is_new":false},{"rank":12,"model_id":"qwen3.5-9b","model_name":"Qwen3.5-9B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.915,"normalized_score":0.915,"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":"2026-03-02","announcement_date":"2026-03-02","multimodal":true,"param_count":9000000000,"is_new":false},{"rank":13,"model_id":"gemma-3-27b-it","model_name":"Gemma 3 27B","organization_name":"Google","organization_id":"google","benchmark_score":0.904,"normalized_score":0.904,"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":"2025-03-12","announcement_date":"2025-03-12","multimodal":true,"param_count":27000000000,"is_new":false},{"rank":14,"model_id":"nvidia-nemotron-nano-9b-v2","model_name":"Nemotron Nano 9B v2","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.903,"normalized_score":0.903,"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":"2025-08-18","announcement_date":"2025-08-18","multimodal":false,"param_count":8900000000,"is_new":false},{"rank":15,"model_id":"gemma-3-4b-it","model_name":"Gemma 3 4B","organization_name":"Google","organization_id":"google","benchmark_score":0.902,"normalized_score":0.902,"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":"2025-03-12","announcement_date":"2025-03-12","multimodal":true,"param_count":4000000000,"is_new":false},{"rank":16,"model_id":"qwen3.5-4b","model_name":"Qwen3.5-4B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.898,"normalized_score":0.898,"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":"2026-03-02","announcement_date":"2026-03-02","multimodal":true,"param_count":4000000000,"is_new":false},{"rank":17,"model_id":"kimi-k2-instruct-0905","model_name":"Kimi K2-Instruct-0905","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.898,"normalized_score":0.898,"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":"2025-09-05","announcement_date":"2025-09-05","multimodal":false,"param_count":1000000000000,"is_new":false},{"rank":18,"model_id":"kimi-k2-instruct","model_name":"Kimi K2 Instruct","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.898,"normalized_score":0.898,"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":"2025-07-11","announcement_date":"2025-07-11","multimodal":false,"param_count":1000000000000,"is_new":false},{"rank":19,"model_id":"nova-lite","model_name":"Nova Lite","organization_name":"Amazon","organization_id":"amazon","benchmark_score":0.897,"normalized_score":0.897,"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":"2024-11-20","announcement_date":"2024-11-20","multimodal":true,"param_count":null,"is_new":false},{"rank":20,"model_id":"longcat-flash-chat","model_name":"LongCat-Flash-Chat","organization_name":"Meituan","organization_id":"meituan","benchmark_score":0.8965,"normalized_score":0.8965,"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":"2025-08-29","announcement_date":"2025-08-29","multimodal":false,"param_count":560000000000,"is_new":false},{"rank":21,"model_id":"exaone-4.5-33b","model_name":"EXAONE 4.5 33B","organization_name":"LG AI Research","organization_id":"lg","benchmark_score":0.896,"normalized_score":0.896,"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":"2026-04-09","announcement_date":"2026-04-09","multimodal":true,"param_count":33000000000,"is_new":false},{"rank":22,"model_id":"llama-3.1-nemotron-ultra-253b-v1","model_name":"Llama 3.1 Nemotron Ultra 253B v1","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.8945,"normalized_score":0.8945,"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":"2025-04-07","announcement_date":"2025-04-07","multimodal":false,"param_count":253000000000,"is_new":false},{"rank":23,"model_id":"qwen3-next-80b-a3b-thinking","model_name":"Qwen3-Next-80B-A3B-Thinking","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.889,"normalized_score":0.889,"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":"2025-09-10","announcement_date":"2025-09-10","multimodal":false,"param_count":80000000000,"is_new":false},{"rank":24,"model_id":"gemma-3-12b-it","model_name":"Gemma 3 12B","organization_name":"Google","organization_id":"google","benchmark_score":0.889,"normalized_score":0.889,"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":"2025-03-12","announcement_date":"2025-03-12","multimodal":true,"param_count":12000000000,"is_new":false},{"rank":25,"model_id":"qwen3-235b-a22b-instruct-2507","model_name":"Qwen3-235B-A22B-Instruct-2507","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.887,"normalized_score":0.887,"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":"2025-07-22","announcement_date":"2025-07-22","multimodal":false,"param_count":235000000000,"is_new":false},{"rank":26,"model_id":"llama-3.1-405b-instruct","model_name":"Llama 3.1 405B Instruct","organization_name":"Meta","organization_id":"meta","benchmark_score":0.886,"normalized_score":0.886,"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":"2024-07-23","announcement_date":"2024-07-23","multimodal":false,"param_count":405000000000,"is_new":false},{"rank":27,"model_id":"qwen3-vl-235b-a22b-thinking","model_name":"Qwen3 VL 235B A22B Thinking","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.882,"normalized_score":0.882,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":236000000000,"is_new":false},{"rank":28,"model_id":"gpt-4.5","model_name":"GPT-4.5","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.882,"normalized_score":0.882,"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":"2025-02-27","announcement_date":"2025-02-27","multimodal":true,"param_count":null,"is_new":false},{"rank":29,"model_id":"qwen3-vl-32b-thinking","model_name":"Qwen3 VL 32B Thinking","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.878,"normalized_score":0.878,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":33000000000,"is_new":false},{"rank":30,"model_id":"qwen3-vl-235b-a22b-instruct","model_name":"Qwen3 VL 235B A22B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.878,"normalized_score":0.878,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":236000000000,"is_new":false},{"rank":31,"model_id":"qwen3-235b-a22b-thinking-2507","model_name":"Qwen3-235B-A22B-Thinking-2507","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.878,"normalized_score":0.878,"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":"2025-07-25","announcement_date":"2025-07-25","multimodal":false,"param_count":235000000000,"is_new":false},{"rank":32,"model_id":"qwen3-next-80b-a3b-instruct","model_name":"Qwen3-Next-80B-A3B-Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.876,"normalized_score":0.876,"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":"2025-09-10","announcement_date":"2025-09-10","multimodal":false,"param_count":80000000000,"is_new":false},{"rank":33,"model_id":"llama-3.1-70b-instruct","model_name":"Llama 3.1 70B Instruct","organization_name":"Meta","organization_id":"meta","benchmark_score":0.875,"normalized_score":0.875,"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":"2024-07-23","announcement_date":"2024-07-23","multimodal":false,"param_count":70000000000,"is_new":false},{"rank":34,"model_id":"gpt-4.1-2025-04-14","model_name":"GPT-4.1","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.874,"normalized_score":0.874,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":2.0,"output_cost_per_million":8.0,"speed_rps":100.0,"context_window":1047576,"release_date":"2025-04-14","announcement_date":"2025-04-14","multimodal":true,"param_count":null,"is_new":false},{"rank":35,"model_id":"kimi-k1.5","model_name":"Kimi-k1.5","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.872,"normalized_score":0.872,"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":"2025-01-20","announcement_date":"2025-01-20","multimodal":true,"param_count":null,"is_new":false},{"rank":36,"model_id":"nova-micro","model_name":"Nova Micro","organization_name":"Amazon","organization_id":"amazon","benchmark_score":0.872,"normalized_score":0.872,"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":"2024-11-20","announcement_date":"2024-11-20","multimodal":false,"param_count":null,"is_new":false},{"rank":37,"model_id":"deepseek-v3","model_name":"DeepSeek-V3","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.861,"normalized_score":0.861,"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":"2024-12-25","announcement_date":"2024-12-25","multimodal":false,"param_count":671000000000,"is_new":false},{"rank":38,"model_id":"qwen3-vl-30b-a3b-instruct","model_name":"Qwen3 VL 30B A3B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.858,"normalized_score":0.858,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":31000000000,"is_new":false},{"rank":39,"model_id":"phi-4-reasoning-plus","model_name":"Phi 4 Reasoning Plus","organization_name":"Microsoft","organization_id":"microsoft","benchmark_score":0.849,"normalized_score":0.849,"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":"2025-04-30","announcement_date":"2025-04-30","multimodal":false,"param_count":14000000000,"is_new":false},{"rank":40,"model_id":"sarvam-105b","model_name":"Sarvam-105B","organization_name":"Sarvam AI","organization_id":"sarvamai","benchmark_score":0.848,"normalized_score":0.848,"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":"2026-03-06","announcement_date":"2026-03-06","multimodal":false,"param_count":105000000000,"is_new":false},{"rank":41,"model_id":"qwen3-vl-32b-instruct","model_name":"Qwen3 VL 32B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.847,"normalized_score":0.847,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":33000000000,"is_new":false},{"rank":42,"model_id":"gpt-4.1-mini-2025-04-14","model_name":"GPT-4.1 mini","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.841,"normalized_score":0.841,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":0.4,"output_cost_per_million":1.6,"speed_rps":150.0,"context_window":1047576,"release_date":"2025-04-14","announcement_date":"2025-04-14","multimodal":true,"param_count":null,"is_new":false},{"rank":43,"model_id":"qwen-2.5-72b-instruct","model_name":"Qwen2.5 72B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.841,"normalized_score":0.841,"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":"2024-09-19","announcement_date":"2024-09-19","multimodal":false,"param_count":72700000000,"is_new":false},{"rank":44,"model_id":"qwq-32b","model_name":"QwQ-32B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.839,"normalized_score":0.839,"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":"2025-03-05","announcement_date":"2025-03-05","multimodal":false,"param_count":32500000000,"is_new":false},{"rank":45,"model_id":"qwen3-vl-8b-instruct","model_name":"Qwen3 VL 8B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.837,"normalized_score":0.837,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":9000000000,"is_new":false},{"rank":46,"model_id":"phi-4-reasoning","model_name":"Phi 4 Reasoning","organization_name":"Microsoft","organization_id":"microsoft","benchmark_score":0.834,"normalized_score":0.834,"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":"2025-04-30","announcement_date":"2025-04-30","multimodal":false,"param_count":14000000000,"is_new":false},{"rank":47,"model_id":"qwen3-vl-8b-thinking","model_name":"Qwen3 VL 8B Thinking","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.832,"normalized_score":0.832,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":9000000000,"is_new":false},{"rank":48,"model_id":"mistral-small-24b-instruct-2501","model_name":"Mistral Small 3 24B Instruct","organization_name":"Mistral AI","organization_id":"mistral","benchmark_score":0.829,"normalized_score":0.829,"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":"2025-01-30","announcement_date":"2025-01-30","multimodal":false,"param_count":24000000000,"is_new":false},{"rank":49,"model_id":"qwen3-vl-4b-thinking","model_name":"Qwen3 VL 4B Thinking","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.826,"normalized_score":0.826,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.1,"output_cost_per_million":1.0,"speed_rps":null,"context_window":262144,"release_date":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":4000000000,"is_new":false},{"rank":50,"model_id":"lfm-2.5-vl-3b","model_name":"LFM2.5-VL-3B","organization_name":"Liquid AI","organization_id":"liquid","benchmark_score":0.823,"normalized_score":0.823,"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":"2026-08-12","announcement_date":"2026-08-12","multimodal":true,"param_count":3123483888,"is_new":false},{"rank":51,"model_id":"qwen3-vl-4b-instruct","model_name":"Qwen3 VL 4B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.823,"normalized_score":0.823,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.1,"output_cost_per_million":0.6,"speed_rps":null,"context_window":262144,"release_date":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":4000000000,"is_new":false},{"rank":52,"model_id":"qwen3-vl-30b-a3b-thinking","model_name":"Qwen3 VL 30B A3B Thinking","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.817,"normalized_score":0.817,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":31000000000,"is_new":false},{"rank":53,"model_id":"gpt-4o-2024-08-06","model_name":"GPT-4o","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.81,"normalized_score":0.81,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":2.5,"output_cost_per_million":10.0,"speed_rps":132.0,"context_window":128000,"release_date":"2024-08-06","announcement_date":"2024-08-06","multimodal":true,"param_count":null,"is_new":false},{"rank":54,"model_id":"llama-3.1-8b-instruct","model_name":"Llama 3.1 8B Instruct","organization_name":"Meta","organization_id":"meta","benchmark_score":0.804,"normalized_score":0.804,"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":"2024-07-23","announcement_date":"2024-07-23","multimodal":false,"param_count":8000000000,"is_new":false},{"rank":55,"model_id":"gemma-3-1b-it","model_name":"Gemma 3 1B","organization_name":"Google","organization_id":"google","benchmark_score":0.802,"normalized_score":0.802,"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":"2025-03-12","announcement_date":"2025-03-12","multimodal":false,"param_count":1000000000,"is_new":false},{"rank":56,"model_id":"llama-3.1-nemotron-nano-8b-v1","model_name":"Llama 3.1 Nemotron Nano 8B V1","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.793,"normalized_score":0.793,"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":"2025-03-18","announcement_date":"2025-03-18","multimodal":false,"param_count":8000000000,"is_new":false},{"rank":57,"model_id":"qwen3.5-2b","model_name":"Qwen3.5-2B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.786,"normalized_score":0.786,"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":"2026-03-02","announcement_date":"2026-03-02","multimodal":true,"param_count":2000000000,"is_new":false},{"rank":58,"model_id":"llama-3.2-3b-instruct","model_name":"Llama 3.2 3B Instruct","organization_name":"Meta","organization_id":"meta","benchmark_score":0.774,"normalized_score":0.774,"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":"2024-09-25","announcement_date":"2024-09-25","multimodal":false,"param_count":3210000000,"is_new":false},{"rank":59,"model_id":"minicpm-sala","model_name":"MiniCPM-SALA","organization_name":"OpenBMB","organization_id":"openbmb","benchmark_score":0.7634,"normalized_score":0.7634,"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":"2026-02-11","announcement_date":"2026-02-11","multimodal":false,"param_count":9477203968,"is_new":false},{"rank":60,"model_id":"north-micro-vision-instruct","model_name":"North Micro Vision Instruct","organization_name":"Cohere","organization_id":"cohere","benchmark_score":0.749,"normalized_score":0.749,"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":"2026-08-12","announcement_date":"2026-08-12","multimodal":true,"param_count":2400000000,"is_new":false},{"rank":61,"model_id":"granite-3.3-8b-base","model_name":"Granite 3.3 8B Base","organization_name":"IBM","organization_id":"ibm","benchmark_score":0.7482,"normalized_score":0.7482,"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":"2025-04-16","announcement_date":"2025-04-16","multimodal":true,"param_count":8170000000,"is_new":false},{"rank":62,"model_id":"granite-3.3-8b-instruct","model_name":"Granite 3.3 8B Instruct","organization_name":"IBM","organization_id":"ibm","benchmark_score":0.7482,"normalized_score":0.7482,"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":"2025-04-16","announcement_date":"2025-04-16","multimodal":true,"param_count":8000000000,"is_new":false},{"rank":63,"model_id":"gpt-4.1-nano-2025-04-14","model_name":"GPT-4.1 nano","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.745,"normalized_score":0.745,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":0.1,"output_cost_per_million":0.4,"speed_rps":200.0,"context_window":1047576,"release_date":"2025-04-14","announcement_date":"2025-04-14","multimodal":true,"param_count":null,"is_new":false},{"rank":64,"model_id":"qwen-2.5-7b-instruct","model_name":"Qwen2.5 7B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.712,"normalized_score":0.712,"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":"2024-09-19","announcement_date":"2024-09-19","multimodal":false,"param_count":7610000000,"is_new":false},{"rank":65,"model_id":"granite-4.0-tiny-preview","model_name":"IBM Granite 4.0 Tiny Preview","organization_name":"IBM","organization_id":"ibm","benchmark_score":0.63,"normalized_score":0.63,"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":"2025-05-02","announcement_date":"2025-05-02","multimodal":false,"param_count":7000000000,"is_new":false},{"rank":66,"model_id":"phi-4","model_name":"Phi 4","organization_name":"Microsoft","organization_id":"microsoft","benchmark_score":0.63,"normalized_score":0.63,"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":"2024-12-12","announcement_date":"2024-12-12","multimodal":false,"param_count":14700000000,"is_new":false},{"rank":67,"model_id":"pixtral-12b-2409","model_name":"Pixtral-12B","organization_name":"Mistral AI","organization_id":"mistral","benchmark_score":0.613,"normalized_score":0.613,"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":"2024-09-17","announcement_date":"2024-09-17","multimodal":true,"param_count":12400000000,"is_new":false},{"rank":68,"model_id":"qwen3.5-0.8b","model_name":"Qwen3.5-0.8B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.44,"normalized_score":0.44,"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":"2026-03-02","announcement_date":"2026-03-02","multimodal":true,"param_count":800000000,"is_new":false}]}