09-13-2026, 04:42 PM
Curious what people here are seeing.
I'm mostly working with DeepSeek in vLLM on M3 Max. The part I'm trying to understand is speculative decoding.
If you've tested something similar, what did you keep and what did you stop bothering with?
I'm mostly working with DeepSeek in vLLM on M3 Max. The part I'm trying to understand is speculative decoding.
If you've tested something similar, what did you keep and what did you stop bothering with?
still benchmarking everything