Unverified50% confidenceFactExact time
Keepace.ai performs post-training based on Qwen3.5-27B, focusing on Chinese exercise guidance.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/30/2026
First Seen
Valid until: 1/28/2027
Sources
Related Entities
Related Claims
UnverifiedBonsai 2 uses a full retraining starting from the Qwen3-27B architecture with ternary constraints as the objective, rather than applying post-training quantization to existing weights.67% similarUnverifiedOpenAI的GPT-4o mini、Anthropic的Claude 3.5 Haiku等策略遵循'一次预训练、多次后训练'的成本摊薄逻辑64% similarUnverifiedAlibaba open-sourced a framework called QwenGyre, aimed at solving the problem of GPUs idling while waiting during long-task agent training, using asynchronous scheduling to let agents complete training while executing long tasks.62% similarUnverifiedThe later stages of Qwen3 training made extensive use of outcome-reward-based reinforcement learning.61% similarUnverifiedQwen3.5 implements a reinforcement learning framework with separated training and inference, capable of supporting a replay system with tens of trillions of tokens.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1114768API
curl https://kongchang.com/api/v1/knowledge/claims/1114768MCP
get_claim(id=1114768)