Tech2Wild
@Tech2Wild
đŽ Tech, gaming, AI, and everything in between. đ¤ Building with it, not just talking about it. đĨ From the mind of @ToNYD2WiLD
Joined March 2026
222 Following 4.8K Followers
GLM 5.3 Flash on 4 DGX Sparks with TensorFold đĨ
vs my vLLM TP4 stack, same harness:
⥠Code 127 vs 120 tok/s
⥠JSON 133 vs 120
đ ī¸ Tool calls 93 vs 68 (+37%)
đ Faster at every concurrency, C1 to C6
đ 500K-token prompt in 191s (vLLM 202s)
đ§ 8 streams, 500K context each, vision on
Show more
