Deepseek 4.1 Flash

Accelerated Computing DGX Station / GB300 / GB200
1

1st run 3tok/sec
Working on optimization

2

Update

Updated; pinning experts; single-stream prose 89.3 → 153.1 tok/s (+71%), eight streams 308 → 701, sixteen streams 414 → 809, shell 160 → 250, code 154 → 214, structured 123 → 208. KV 4.75 GiB (was 4.89)