Expired50% confidenceFactExact time
Coin VLA is based on Coin 3.54B (3.54 billion parameters) as its backbone, belonging to a lightweight architecture.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
XAI Releases Grok Build Programming Model, Gemini Spark Opens to Ultra Users
bilibiliinfinite灵感港5/30/2026
Related Claims
UnverifiedV4-Pro uses a Mixture of Experts (MoE) architecture with a total parameter count of 1.6 trillion (1.6T) and 49 billion (49B) active parameters.60% similarUnverifiedMeta's LLaMA 3 series has a maximum scale of 405B, and DeepSeek-V3 is approximately 671B (mixture-of-experts architecture).58% similarUnverifiedVGGNet由牛津大学视觉几何组在2014年提出，其核心设计是使用3×3卷积核堆叠替代大卷积核58% similarUnverifiedvLLM 支持多 GPU 分布式推理，主要采用张量并行策略，底层集成 Megatron-LM 的并行算子并利用 NCCL 库实现 GPU 间通信54% similarUnverifiedRDNA 3.5 is a GPU microarchitecture launched by AMD in 2024, primarily targeting the integrated graphics of the Ryzen AI 300 series (codenamed Strix Point).54% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/51226API
curl https://kongchang.com/api/v1/knowledge/claims/51226MCP
get_claim(id=51226)