Home Model series Qwen Qwen · Terminal-Bench 2.0 · Version / Unit not specified · Release history Qwen · Terminal-Bench 2.0 · Version / Unit not specified · Release history View scores, configurations, and disclosures for Terminal-Bench 2.0 · Version / Unit not specified across 2 Qwen releases, ordered by release date.
Differences are calculated only for unique, active, exact values between direct predecessors when the unit, metric, and explicitly recorded configuration and conditions match.
2026-04-16 · 1 score record for this release, including 1 exact value (may include historical records)
Qwen3.6-35B-A3B · Terminal-Bench 2.0 scores and configurations Metric and score Configuration and evaluation conditions Sources Reported score (unit unspecified) 51.5 Reported 51.5 · Unit not specified Published Unconfirmed
Qwen3.6-35B-A3B (reported) Settings 官方 Qwen3.6-35B-A3B 视觉语言 post-trained 开放权重;表格未说明逐项实测模式;不据型号之外材料补充量化或服务配置。 Unknowns 模型卡称 Qwen3.6 默认运行 thinking mode;每项成绩的完整解码与工具设置仅在明确脚注时记录。、默认支持 thinking 不证明表格逐项实测模式;成绩未明确标注思考开关或 effort 时保持未注明。 Tools Tool access not recorded Reasoning budget Output limit 80K tokens Framework Harbor/Terminus-2 Protocol Harbor/Terminus-2 Conditions * Terminal-Bench 2.0: Harbor/Terminus-2 harness; 3h timeout, 32 CPU/48 GB RAM; temp=1.0, top_p=0.95, top_k=20, max_tokens=80K, 256K ctx; avg of 5 runs. View this score in rankings
2026-04-22 · 1 score record for this release, including 1 exact value (may include historical records)
Qwen3.6-27B · Terminal-Bench 2.0 scores and configurations Metric and score Configuration and evaluation conditions Sources Reported score (unit unspecified) 59.3 Reported 59.3 · Unit not specified Published Unconfirmed
Qwen3.6-27B (reported) Settings 官方 Qwen3.6-27B 视觉语言 post-trained 开放权重;表格未说明逐项实测模式;模型卡 Model Overview 列明 27B 参数。 Unknowns 模型卡称 Qwen3.6 默认运行 thinking mode;每项成绩的完整解码与工具设置仅在明确脚注时记录。、默认支持 thinking 不证明表格逐项实测模式;成绩未明确标注思考开关或 effort 时保持未注明。 Tools Tool access not recorded Reasoning budget Output limit 80K tokens Framework Harbor/Terminus-2 Protocol Harbor/Terminus-2 Conditions * Terminal-Bench 2.0: Harbor/Terminus-2 harness; 3h timeout, 32 CPU/48 GB RAM; temp=1.0, top_p=0.95, top_k=20, max_tokens=80K, 256K ctx; avg of 5 runs. View this score in rankings