Frontier Log

AlibabaModel

Qwen3.8-Flash-Next

Made by
Alibaba
Released
Announced in
Qwen3.8-Flash-Next: A New Architecture, Towards Ultimate Cost-Efficiencyqwen.ai
Event
Qwen3.8-Flash-Next
API record
/api/v1/models/qwen3-8-flash-next

Benchmark results

BenchmarkScoreSource check
Humanity's Last ExamNo tools · GPT-4o judge; full or text-only set not stated35.9%
CorroboratedThe same number is printed in 1 other lab document: 1
SWE-bench ProClaude Code harness; 256K context · Run on the lab's own corrected version of the public set62.5%
CorroboratedThe same number is printed in 1 other lab document: 1
GPQA Diamond91.7%
CorroboratedThe same number is printed in 1 other lab document: 1
DeepSWE v1.1the higher of the Claude Code and mini-SWE-agent harnesses; 256K context58.7%
CorroboratedThe same number is printed in 1 other lab document: 1
Sources

Reported by Alibaba

Events

All models