MixTAO-7Bx2-MoE-v8.1 (Publisher: zhengr): benchmark results

Provider: Other. Released 2024-02-26. Access: Open.

Aggregate Rating 1423 ± 43, rank #5017 of 10916 rated models, from 14 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open Chinese LLM - TruthfulQA MC64.5Accuracy (%)96
Open Chinese LLM - WinoGrande65.04Accuracy (%)70.7
Open Chinese LLM - HellaSwag61.49Accuracy (%)63.3
Open Chinese LLM - ARC Challenge53.33Accuracy (%)55.6
Open LLM Leaderboard - GPQA29.87Score54.5
Open Chinese LLM - GSM8K46.32Accuracy (%)53.1
Open Chinese LLM Leaderboard57.31Official Nonzero-Task Mean (%)53
Open LLM Leaderboard - IFEval41.88Score42.7
Open LLM Leaderboard - MuSR39.76Score39.9
Open LLM Leaderboard - MATH Level 56.04Score33
Open Chinese LLM - C-Eval Semantic63.12Accuracy (%)32.4
Open LLM Leaderboard - MMLU-Pro28.47Score32.1

Interactive version: theaggregate.ai/model?slug=mixtao-7bx2-moe-v8-1-publisher-zhengr · How It Works · Data refreshed daily, snapshot 2026-11-08.