Hugging Face Official Benchmarks: Daily Update 2026-10-20

A public leaderboard map tracking Hugging Face official benchmarks reported 49 benchmarks and 1,107 leaderboard entries as of October 20, 2026. FINAL-Bench's Darwin-180B-RSI held the most top spots with 11, including MMLU-Pro (88.12), GPQA (94.44), AIME 2026 (100) and HMMT Feb 2026 (100), while deepseek-ai's DeepSeek-V4.1-Flash led HLE (63.9) and terminal-bench-2.1 (90.6). inclusionAI, zai-org, XiaomiMiMo and deepseek-ai followed in the organization rankings.

This is the daily update of the Hugging Face Official Benchmarks map https://huggingface.co/spaces/quantid/huggingface-official-benchmark-leaderboards , which reads every Hugging Face official benchmark leaderboard straight from the public API. As of 2026-10-20 KST : 49 official benchmarks, 1,107 leaderboard entries. | Rank | Organization | 1 spots | |---|---|---| | 1 | FINAL-Bench | 11 | | 2 | inclusionAI | 4 | | 3 | zai-org | 4 | | 4 | XiaomiMiMo | 3 | | 5 | deepseek-ai | 3 | | 6 | MiniMaxAI | 2 | | 7 | moonshotai | 2 | | 8 | ornith-ai | 2 | | 9 | Edge0 | 1 | | 10 | KDLAI | 1 | | Benchmark | 1 model | Score | Entries | |---|---|---|---| | MMLU-Pro https://huggingface.co/datasets/TIGER-Lab/MMLU-Pro | FINAL-Bench/Darwin-180B-RSI | 88.12 | 142 | | gpqa https://huggingface.co/datasets/Idavidrein/gpqa | FINAL-Bench/Darwin-180B-RSI | 94.44 | 115 | | hle https://huggingface.co/datasets/cais/hle | deepseek-ai/DeepSeek-V4.1-Flash | 63.9 | 95 | | SWE-bench Verified https://huggingface.co/datasets/SWE-bench/SWE-bench Verified | ornith-ai/Ornith-1.5-397B | 86 | 68 | | open-asr-leaderboard https://huggingface.co/datasets/hf-audio/open-asr-leaderboard | Edge0/ARK-ASR-3B | 4.76 | 63 | | WildClawBench https://huggingface.co/datasets/internlm/WildClawBench | Qwen/Qwen3.8-2.4T-A95B | 708 | 51 | | SWE-bench Pro https://huggingface.co/datasets/ScaleAI/SWE-bench Pro | inclusionAI/Ling-3.1-flash | 68.74 | 47 | | terminal-bench-2.0 https://huggingface.co/datasets/harborframework/terminal-bench-2.0 | mindlab-research/Macaron-V1-Venti | 87.6 | 40 | | terminal-bench-2.1 https://huggingface.co/datasets/harborframework/terminal-bench-2.1 | deepseek-ai/DeepSeek-V4.1-Flash | 90.6 | 37 | | ParseBench https://huggingface.co/datasets/llamaindex/ParseBench | KDLAI/KDL-Frontier-Parser-nano | 76.36 | 34 | | SWE-bench Multilingual https://huggingface.co/datasets/SWE-bench/SWE-bench Multilingual | inclusionAI/Ling-3.1-flash | 88.5 | 28 | | aime 2026 https://huggingface.co/datasets/MathArena/aime 2026 | FINAL-Bench/Darwin-180B-RSI | 100 | 27 | | deep-swe https://huggingface.co/datasets/datacurve/deep-swe | deepseek-ai/DeepSeek-V4.1-Flash | 74.2 | 25 | | ScreenSpot-Pro https://huggingface.co/datasets/likaixin/ScreenSpot-Pro | meta-models/Muse-Glimmer-30B | 75.4 | 24 | | Claw-Eval https://huggingface.co/datasets/claw-eval/Claw-Eval | ornith-ai/Ornith-1.5-397B | 81.4 | 23 | | olmOCR-bench https://huggingface.co/datasets/allenai/olmOCR-bench | SoMarkAI/OXR-1.0 | 87.84 | 22 | | hmmt feb 2026 https://huggingface.co/datasets/MathArena/hmmt feb 2026 | FINAL-Bench/Darwin-180B-RSI | 100 | 20 | | gsm8k https://huggingface.co/datasets/openai/gsm8k | XiaomiMiMo/MiMo-V2.5-Pro | 99.6 | 19 | | Toolathlon https://huggingface.co/datasets/hkust-nlp/Toolathlon | XiaomiMiMo/MiMo-V2.6-Pro-RL | 76.9 | 18 | | Real5-OmniDocBench https://huggingface.co/datasets/PaddlePaddle/Real5-OmniDocBench | PaddlePaddle/PaddleOCR-VL-1.6 | 93.19 | 17 | | MDPBench https://huggingface.co/datasets/Delores-Lin/MDPBench | FINAL-Bench/Darwin-180B-RSI | 83.65 | 15 | | RedlineBench https://huggingface.co/datasets/crosbylegal/RedlineBench | crosbylegal/gemini-4-argon | 59.5 | 14 | | apex-agents https://huggingface.co/datasets/mercor/apex-agents | moonshotai/Kimi-K3 | 41 | 13 | | skillsbench https://huggingface.co/datasets/benchflow/skillsbench | inclusionAI/Ling-3.1-flash | 68.7 | 12 | | LEXam-hard https://huggingface.co/datasets/joelniklaus/LEXam-hard | FINAL-Bench/Darwin-180B-RSI | 45.72 | 12 | | ExtractBench https://huggingface.co/datasets/llamaindex/ExtractBench | FINAL-Bench/Darwin-180B-RSI-R3 | 90.29 | 11 | | MMMU Pro https://huggingface.co/datasets/MMMU/MMMU Pro | FINAL-Bench/Darwin-180B-RSI | 79.48 | 11 | | LEXam https://huggingface.co/datasets/LEXam-Benchmark/LEXam | FINAL-Bench/Darwin-180B-RSI | 68.94 | 10 | | yc-bench https://huggingface.co/datasets/collinear-ai/yc-bench | MiniMaxAI/MiniMax-M3 | 2100000 | 10 | | Vaani-Benchmark-V1.0 https://huggingface.co/datasets/ARTPARK-IISc/Vaani-Benchmark-V1.0 | gnani/gnani | 13 | 10 | | Long-Horizon-Terminal-Bench https://huggingface.co/datasets/IntelligenceLab/Long-Horizon-Terminal-Bench | moonshotai/Kimi-K3 | 6 | 9 | | typed-decisions https://huggingface.co/datasets/LocalLLaMA/typed-decisions | FINAL-Bench/Darwin-27B-ZTC-v5 | 0.812 | 8 | | agents-last-exam https://huggingface.co/datasets/agents-last-exam/agents-last-exam | XiaomiMiMo/MiMo-V2.6-Pro-RL | 31.6 | 8 | | WBench https://huggingface.co/datasets/meituan-longcat/WBench | robbyant/lingbot-world-fast | 78.8 | 8 | | ResearchClawBench https://huggingface.co/datasets/InternScience/ResearchClawBench | zai-org/GLM-5.2 | 20.709230769230768 | 7 | | Video-MME-v2 https://huggingface.co/datasets/MME-Benchmarks/Video-MME-v2 | MiniMaxAI/MiniMax-M3 | 85.4 | 7 | | ifstruct-v1.0 https://huggingface.co/datasets/LiquidAI/ifstruct-v1.0 | FINAL-Bench/Darwin-180B-RSI | 98.95 | 6 | | terminal-bench https://huggingface.co/datasets/harborframework/terminal-bench | inclusionAI/Ling-3.1-flash | 46.03 | 6 | | EvasionBench https://huggingface.co/datasets/FutureMa/EvasionBench | zai-org/GLM-4.7 | 82.91 | 5 | | chi-bench https://huggingface.co/datasets/actava/chi-bench | zai-org/GLM-5.1 | 18.7 | 3 | | arguana https://huggingface.co/datasets/mteb/arguana | google/embeddinggemma-300m | 71.535 | 3 | | APEX-v1-extended https://huggingface.co/datasets/mercor/APEX-v1-extended | zai-org/GLM-4.7 | 51.7 | 2 | | vlabench primitive ft lerobot video https://huggingface.co/datasets/VLABench/vlabench primitive ft lerobot video | nvidia/GR00T-N1-2B | 39.7 | 1 | | BRIGHT https://huggingface.co/datasets/mteb/BRIGHT | mteb/baseline-bm25s | 0.11 | 1 | Most leaderboard entries are self-reported by model authors, and evaluation settings differ between submissions sampling, voting, thinking budget . An entry is marked verified only where Hugging Face marks it so. Ties at rank 1 count for each tied organization. Disclosure: QuantID is a technology partner of VIDRAFT, whose FINAL-Bench organization appears in these rankings. This post reports the public leaderboard data as published on Hugging Face; it is not an independent evaluation.