Weighted / imatrix quants of aimeri/spoomplesmaxx-magpie-35B-A3.

Static quants are not provided: at every size that matters the imatrix quant of the same name is at least as good, so this repo is the whole set.

The in-domain half exists because this is a sparse MoE: a token only touches its routed experts, so a small generic corpus leaves rare experts with no importance data at all. Coverage was gated before quantizing — every quantizable tensor has importance data and every expert row was activated above a token floor.

The Q4_K_M quant was verified after quantization: greedy generations terminate on <|im_end|> (not the token cap), think blocks close, no stray glyphs.

KL divergence of each quant against this model's own bf16, on a holdout that was excluded from calibration (evaluating on your own calibration domain flatters the numbers). Lower KLD is closer to the unquantized model.

same top-1 is the share of holdout positions where the quant's most likely token is unchanged from bf16.

Roughly, higher in the table is smaller and worse. i1-Q4_K_M is the recommended default; i1-IQ4_XS if you need to save a little; i1-Q6_K if you have the memory and want to stop thinking about it.

--jinja matters: the chat template is embedded in the GGUF and pre-opens the <think> block, which is the behaviour the model was trained for. Tool calls use Qwen3.5's XML convention (<tool_call><function=NAME><parameter=KEY>VALUE</parameter></function></tool_call>).