Qwen3.6-35B-A3B-oQ4e-mtp
[!NOTE] Updated 2026-10-29: Re-quantized with the improved oQe quantization in oMLX 0.7.3. Please re-download if you have an earlier copy.
This model was quantized using oQ (oMLX v0.7.3) mixed-precision quantization with improved oQe (jundot/omlx#4930).
A mixed-precision MLX checkpoint of Qwen/Qwen3.6-35B-A3B, a mixture-of-experts vision-language model with 35B total and 3B activated parameters and a hybrid layout of Gated DeltaNet and Gated Attention layers. It includes the language model, the vision encoder and the multi-token prediction (MTP) head.
The effective average is 4.786 bits/weight for the language backbone and 4.925 bits/weight for the complete checkpoint.
Quantization and Bit Distribution
oQ4e allocates bits by measured layer sensitivity, then rounds each group with an importance matrix (128 samples x 512 tokens) and a weighted least-squares refit of its scale and bias.
| Storage format | Logical weights | Share of total | Tensor storage | Effective bits/weight |
|---|---|---|---|---|
| Affine 4-bit, group 64 | 33.018B | 91.838% | 17.297 GiB | 4.50 |
| Affine 8-bit, group 128 | 0.129B | 0.359% | 0.124 GiB | 8.25 |
| Affine 8-bit, group 64 | 2.328B | 6.474% | 2.303 GiB | 8.50 |
| BF16 | 0.478B | 1.329% | 0.890 GiB | 16.00 |
| Total | 35.952B | 100% | 20.614 GiB | 4.925 |
Storage Breakdown
| Component | Logical weights | Storage |
|---|---|---|
| Language backbone | 34.661B | 20.738 GB / 19.313 GiB |
| Vision encoder | 0.447B | 0.893 GB / 0.832 GiB |
| MTP head | 0.845B | 0.503 GB / 0.468 GiB |
| Total | 35.952B | 22.134 GB / 20.614 GiB |
License
Apache-2.0, following the source model. See LICENSE.