Qwen3.8-Flash-Next · 模型资料

相关动态

订阅更新

Unsloth发布Qwen3.8-Flash-Next GGUF2026-11-06T14:11:34.825000+00:00
Qwen3.8-Flash-Next在Nvidia GPU上实现高Token输出2026-10-14T03:43:40.238000+00:00
Cohere报告Qwen3.8-Flash-Next异常行为2026-10-26T16:12:22.214000+00:00

公开成绩与计算依据

综合参考分根据已公开的评测资料整理。查看发布日期依据

综合参考分 76.8 · 12项基准

推理求解 71.0

GPQA Diamond · 0.923232323232323 比例

Artificial Analysis · 独立评测 · Diamond

Qwen3.8-Flash-Next · Current AA standalone protocol. Status verified in aa-method-current.md section gpqa-diamond

HLE · 0.163 比例

Artificial Analysis · 独立评测 · Diamond

Qwen3.8-Flash-Next · Artificial Analysis current published protocol; hle-diamond

工具与任务执行 78.4

tau3-Banking · 0.45360824742268 比例

Artificial Analysis · 独立评测 · 1.0.1

Qwen3.8-Flash-Next · Current AA standalone protocol. Status verified in aa-method-current.md section tau3-banking

GDPval-AA · 1635.78 分

Artificial Analysis · 独立评测 · 2.1

Qwen3.8-Flash-Next · Artificial Analysis current published protocol; gdpval-aa-normalized

Terminal-Bench · 0.252525252525253 比例

Artificial Analysis · 独立评测 · 4.0

Qwen3.8-Flash-Next · Artificial Analysis current published protocol; terminalbench-4-0

AutomationBench · 0.559083218951688 比例

Artificial Analysis · 独立评测 · current published task suite

Qwen3.8-Flash-Next · Artificial Analysis current published protocol; automationbench-aa

AA-Briefcase · 1573.51 分

Artificial Analysis · 独立评测 · 1.1

Qwen3.8-Flash-Next · Artificial Analysis current published protocol; aa-briefcase

编程 100.0

DeepSWE · 58.7

huggingface.co · 官方发布比较 · 1.1

Qwen3.8-Flash-Next · Published page qwen38-flashnext; row Agentic coding DeepSWE 1.1; preserve page footnotes and task-specific settings · 2026-08-26公开

语言与信息处理 66.5

IFBench · 81.3

huggingface.co · 官方发布比较 · current published task suite

Qwen3.8-Flash-Next · Published page qwen38-flashnext; row Instruction following IFBench; preserve page footnotes and task-specific settings · 2026-08-26公开

GDP.pdf · 0.156 比例

Artificial Analysis · 独立评测 · current published task suite

Qwen3.8-Flash-Next · Artificial Analysis current published protocol; gdp-pdf

AA-Omniscience · 0.23533333333333334 比例

Artificial Analysis · 独立评测 · 1.1

Qwen3.8-Flash-Next · Artificial Analysis current published protocol; omniscience-accuracy

AA-Omniscience · 0.532258064516129 比例

Artificial Analysis · 独立评测 · 1.1

Qwen3.8-Flash-Next · Artificial Analysis current published protocol; omniscience-non-hallucination