DeepSeek V4.1 Flash 552B (16B active)
M3 Ultra 256GB · Unsloth Studio
- generation:
- 40.0 tokens/s
- quant:
- IQ3_XXS (GGUF)
User reports DeepSeek V4.1 Flash at 40 t/s output on a Mac Studio with 256GB, run through Unsloth at IQ3_XXS. The user finds it slow in practice, saying it takes 5-10 minutes to finish plotting a map that GLM 5.3 Flash finishes in a minute or so. They ask whether others use speculative decoding and what effort level to set.