0xSero (@0xSero)
@0xSero
0 正在关注 0 粉丝
Inkling 2 Small on 4x RTX Pro 6000s
- 16x 380,000 kv cache concurrency
- 224 tok/s decode on c1
- 8150 tok/s prefill
- exl3-4bit 97.5% top agreement w bf16
Long Live American Open Source.
I love you Thinking Machines, thanks for providing the West still has some spine.
显示更多
Ben was right, GPT-5.5 no thinking is a beast. Holy moly, it's going around faster than I've ever seen doing precise actions to get servers working etc..
5.8k tokens to fix like 20 issues lol.
显示更多
