Unverified50% confidenceReleaseExact time
SuperGrok subscribers can use Grok models in Zed's Agent Panel
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/4/2026
First Seen
Valid until: 1/2/2027
Sources
Related Entities
Related Claims
UnverifiedQwen2.5后训练首次在在线阶段使用GRPO，GRPO是DeepSeek团队提出的PPO轻量化变体，去掉独立价值网络，用组内平均奖励作为基线61% similarUnverifiedThe Grok series emphasizes deep integration with the X platform and the ability to access real-time information.59% similarVerifiedGRPO (Group Relative Policy Optimization), proposed by DeepSeek, requires no separate training of a value network and has become the mainstream algorithm choice for current agentic RL training.58% similarVerifiedGRPO由DeepSeek团队系统化提出并在训练DeepSeek-R1系列模型时得到大规模验证58% similarUnverifiedDeepSeek explored the GRPO algorithm as an attempt to improve upon the original RLHF framework.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/971257API
curl https://kongchang.com/api/v1/knowledge/claims/971257MCP
get_claim(id=971257)