Unverified50% confidenceReleaseExact time
The GRPO training in this update adds support for Qwen3.5 and follows the latest TRL and vLLM.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/18/2026
First Seen
Valid until: 12/17/2026
Sources
Related Entities
Related Claims
UnverifiedGrok 4.6 is an iterative upgrade (dot upgrade) based on Grok 4.5, not a newly trained model.67% similarUnverifiedThe base model is Qwen3-14B, from which two versions—KGModel and CGModel—are trained separately via supervised fine-tuning (SFT).58% similarUnverifiedMainstream LLM products such as ChatGPT, DeepSeek, Kimi, and Qwen have all gone through three core stages: pre-training, SFT, and preference alignment.57% similarUnverified主流量化方案包括GPTQ、AWQ和GGUF，开发者可根据实际场景灵活选择来部署Qwen3.6-27B57% similarUnverified成熟的开发者应同时掌握国内的DeepSeek、Qwen以及国外的OpenAI GPT-4o、o3、Claude 3.5/3.7/4.0等多个模型55% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/932844API
curl https://kongchang.com/api/v1/knowledge/claims/932844MCP
get_claim(id=932844)