Unverified50% confidenceBenchmarkExact time
In a GPU physics simulation task of colliding galaxies with more than 500,000 stars, Argon wrote its own CUDA kernel, first testing a single galaxy to confirm that energy drift was near zero, then running the full collision, achieving nearly 700 billion gravitational calculations per second; after the first rendered video was corrupted, the agent checked on its own and automatically re-rendered it.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/14/2026
First Seen
Valid until: 1/12/2027
Sources
Related Entities
Related Claims
Unverified配备NVIDIA GPU并使用CUDA加速,7B模型可以达到50-100+ token/s的推理速度67% similarUnverifiedIn a harness with terminal access to a local DGX Spark GPU node, Argon first reproduced a research paper on growing neural cellular automata, implemented an exact 8,000-parameter model in PyTorch, trained it on the GPU, cut off half of the organism's body to verify self-healing, and then ported the trained weights to an interactive WebGL shader, taking 23 minutes total from paper to a usable browser demo, with the browser shader and the PyTorch model matching to five decimal places.66% similarUnverifiedGPU显存并非100%可用,因为GPU驱动、CUDA运行时上下文、显示输出和操作系统图形合成器都会预先占用一部分显存64% similarUnverifiedMatrix multiplication on GPUs is computed in parallel across thousands of CUDA cores via the cuBLAS library.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1030354API
curl https://kongchang.com/api/v1/knowledge/claims/1030354MCP
get_claim(id=1030354)