Unverified50% confidenceBenchmarkExact time
Flash 1017 scores 68.9% on LongBench QA top-5, higher than DeepSeek V4-Pro's 66.1% and GPT-4-Turbo's 62.3%.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/17/2026
First Seen
Valid until: 10/31/2026
Sources
Related Entities
Related Claims
UnverifiedDeepSeek's model V4 Flash 0731 achieved a score of 82.7% on Terminal-Bench 2.1.77% similarUnverifiedV4.1-Flash-1017 scores 77.0 on the DeepSWE benchmark.71% similarUnverified盘古2.0 Flash在GPQA Diamond科学推理测试中（Thinking模式）得分83.7，低于千问3.6 35B的86、27B稠密模型的87.8和DeepSeek V4 Flash的88.169% similarUnverifiedFlash 1017 inference depends on DeepSeek's Harness runtime, with optional TensorRT acceleration kernels; the Harness runtime is optimized for A100-class GPUs, and performance drops noticeably on older cards.69% similarUnverifiedIn the black hole demo project, DeepSeek V4.1 Flash performed the best and received 90 points, while GLM5.3 Flash came in last.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1054137API
curl https://kongchang.com/api/v1/knowledge/claims/1054137MCP
get_claim(id=1054137)