DeepSeek-V4.1-Flash-oQ4e-mtp

This model was quantized using oQ (oMLX v0.7.1) mixed-precision quantization with Improved oQe Quantization (jundot/omlx#4385).

A mixed-precision MLX checkpoint of deepseek-ai/DeepSeek-V4.1-Flash, a multimodal mixture-of-experts model with 552B backbone parameters, a Causal Encoder-Decoder layout, Compressed Sparse Attention 2 (CSA2), Engram conditional memory and DSpark speculative decoding. It includes the language model, the Engram tables, the vision encoder, the multimodal aligner and the DSpark MTP weights.

The effective average is 4.365 bits/weight excluding Engram, or 4.529 bits/weight for the complete checkpoint.

Quantization and Bit Distribution

oQ4e keeps the original FP4 and FP8 projection precision, which already fits the budget, and quantizes only the Engram tables.

Storage format	Logical weights	Share of total	Tensor storage	Effective bits/weight
MXFP4, group 32	557.171B	73.004%	275.669 GiB	4.25
MXFP8, group 32	5.959B	0.781%	5.723 GiB	8.25
BF16	3.419B	0.448%	6.369 GiB	16.00
FP32	0.042B	0.006%	0.158 GiB	32.00
Engram affine 4-bit, group 32	196.614B	25.762%	114.444 GiB	5.00
Total	763.205B	100%	402.363 GiB	4.529

Storage Breakdown

Component	Logical weights	Storage
Language backbone	551.881B	300.132 GB / 279.520 GiB
Engram tables	196.614B	122.884 GB / 114.444 GiB
Vision encoder	0.485B	0.971 GB / 0.904 GiB
MTP head	14.225B	8.047 GB / 7.495 GiB
Total	763.205B	432.033 GB / 402.363 GiB

License

MIT, following the source model. See LICENSE.