Unverified50% confidenceReleaseExact time
Thinking Machines released the Dimuon optimizer, a Muon variant designed specifically for LoRA fine-tuning, and it has been released in Tinker.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/16/2026
First Seen
Valid until: 1/14/2027
Sources
Related Entities
Related Claims
UnverifiedThinking Machines Lab has released its first open-weight model, Inkling.71% similarPartially VerifiedMindIE是华为昇腾生态的推理引擎，专为昇腾NPU硬件优化66% similarUnverified昇腾配套的软件栈为CANN（Compute Architecture for Neural Networks），提供与PyTorch、MindSpore等框架的适配层64% similarUnverifiedSoftware engineer Steven Dio developed a Lean checker called NanoLean, used to check whether existing Lean code is correct.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1044472API
curl https://kongchang.com/api/v1/knowledge/claims/1044472MCP
get_claim(id=1044472)