AI 导读
有人把 DeepSeek V4 Pro 0813 与一个叫 J-Space 的 harness 结合了起来，据说能减少扩展推理/工具调用过程中的错误。 在展示的基准测试中，DeepSeek 的表现有些情况下显著提升： Terminal Bench 87.9 → 90.1， NL2Repo 61.5 → 73.4， CyberGym 83.3 → 86.8， DeepSWE 62.7 → 72.0，以及 Toolathlon 74.1 → 79.5。 这一组合在多项智能体/编程基准上也超过了 Fable 5 和 Opus 4.8。
正文
Someone has combined DeepSeek V4 Pro 0813 with a harness called J-Space, which apparently reduces errors in the extended reasoning/tool-use process.
In the benchmarks shown, DeepSeek's performance improves significantly in some cases:
Terminal Bench 87.9 → 90.1,
NL2Repo 61.5 → 73.4,
CyberGym 83.3 → 86.8,
DeepSWE 62.7 → 72.0, and
Toolathlon 74.1 → 79.5.
This combination also outperforms Fable 5 and Opus 4.8 in several agentic/coding benchmarks.
来源：@kimmonismus · x.com