DPT Large — BonfyreFPQ v12
Native .fpq v12 quantized weights for Intel/dpt-large.
Model Info
| Property | Value |
|---|---|
| Architecture | Monocular Depth Estimation |
| Parameters | 344M |
| Original Size | 1304 MB (safetensors) |
| FPQ v12 Size | 329 MB |
| Compression | 4.0× |
| Avg Cosine | 0.999742 |
| Worst Cosine | 0.998460 |
Format
BonfyreFPQ v12 native format: rANS entropy-coded E8 lattice coordinates with 6-bit packed tiles and FP16 scales.
Decode to safetensors
bonfyre-fpqx decode dpt-large-v12-fpq.fpq output.safetensors
Source
Converted from Intel/dpt-large using BonfyreFPQ v12.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support