Ornith-1.5-122B oQe MTP
Collection
8 items • Updated
Ornith-1.5-122B-oQ5e-mtp
This model was quantized using oQ (oMLX v0.7.1) mixed-precision quantization.
The effective average is 5.623 bits/weight for the complete checkpoint.
Quantization and Bit Distribution
| Storage format | Logical weights | Share of total | Tensor storage | Effective bits/weight |
|---|---|---|---|---|
| Affine 5-bit, group 128 | 0.906B | 0.724% | 0.554 GiB | 5.25 |
| Affine 5-bit, group 64 | 118.380B | 94.639% | 75.797 GiB | 5.50 |
| Affine 6-bit, group 64 | 2.780B | 2.222% | 2.104 GiB | 6.50 |
| Affine 8-bit, group 128 | 0.462B | 0.370% | 0.444 GiB | 8.25 |
| Affine 8-bit, group 64 | 2.047B | 1.637% | 2.026 GiB | 8.50 |
| BF16 | 0.511B | 0.408% | 0.951 GiB | 16.00 |
| Total | 125.086B | 100% | 81.876 GiB | 5.623 |
Storage Breakdown
| Component | Logical weights | Storage |
|---|---|---|
| Language model | 122.112B | 85.237 GB / 79.383 GiB |
| Vision | 0.451B | 0.903 GB / 0.841 GiB |
| MTP head | 2.524B | 1.774 GB / 1.652 GiB |
| Total | 125.086B | 87.913 GB / 81.876 GiB |
- Downloads last month
- 144
Model size
125B params
Tensor type
U32
·
BF16 ·
Hardware compatibility
Log In to add your hardware
5-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for scottlowry/Ornith-1.5-122B-oQ5e-mtp
Base model
ornith-ai/Ornith-1.5-122B