It's basically the same as with the Qwen2.5 and 3 series but this time with 1M context and 200k native, yay :)
It's basically the same as with the Qwen2.5 and 3 series but this time with 1M context and 200k native, yay :)