3 ms·
Spec decoding only depends on the tokenizer used. It's transfering either the draft token sequence or at most draft logits to the main model.
by mmoskal 2y ago
Spec decoding only depends on the tokenizer used. It's transfering either the draft token sequence or at most draft logits to the main model.
- jasonjmcghee 2y agoI suppose that makes sense, for some reason I was under the impression that the models need to be aligned / have the same tuning or they'd have different probability distributions and would reject the draft model really often.
- jasonjmcghee 2y agoCould be an lm studio thing, but the qwen3-0.6B model works as a draft model for the qwen3-32B and qwen3-30B-A3B but not the qwen3-235B-A22B model