---
license: apache-2.0
base_model: vllm-sr/d3-lite
library_name: coreml
pipeline_tag: zero-shot-classification
tags:
- coreml
- apple-silicon
- decision-model
- vision
- video
---

# d3-lite — Core ML

[vllm-sr/d3-lite](https://huggingface.co/vllm-sr/d3-lite) (0.85B, part of [Decision 3.0](https://huggingface.co/collections/vllm-sr/decision-30) by the vLLM Semantic Router team) converted to Core ML for Apple silicon: text, image and video requests on device. Same contract as upstream: one forward pass per question, the 255-way FP32 readout at the last prompt token, a probability for every option, no text generated.

## Files

| File | What |
|---|---|
| `d3_E256` / `d3_E640` / `d3_E1536` / `d3_E4096.mlpackage` | text graph (fp16 compute, int8 weights) per prompt-length bucket (prompts up to 4,096 tokens); input = token embeddings with image/video tokens spliced in |
| `vision_P320` / `vision_P1536` / `vision_P3072.mlpackage` | vision tower (fp32) per patch bucket; video frames go through it one frame pair at a time (block mask) |
| `embeddings.f16`, `pos_embed.f32`, `readout.f32`, `tokenizer.json`, `config.json`, `coreml_config.json` | host-side tables and config |
| `python/` | Python host (coremltools): `d3_coreml_text.py`, `d3_coreml_image.py`, `d3_coreml_video.py` |

Swift: `D3VisionManager` in [FluidUse](https://github.com/FluidInference/FluidUse) loads this folder.

## Checked against upstream

Same weights as upstream v3.1.0. Answers that differ from the upstream PyTorch runtime (bf16, MPS), same requests:

| | flips |
|---|---|
| Text (typed-decisions TEST, 20 requests) | 2 / 100 |
| Images (receipt checks) | 1 / 24 |
| Video (4-second street clips, 2 fps) | 0 / 28 |

Speed on an M5 Pro (24 GB), `CPU_AND_GPU`: ~46 ms per text question, 0.32 s per 4-second video look. 

## Python

```python
from d3_coreml_text import D3CoreMLText      # d3_coreml_image.D3CoreMLImage, d3_coreml_video.D3CoreMLVideo
m = D3CoreMLText("<vllm-sr/d3-lite snapshot>", "<this repo>")
m.system_one(state="...", questions={"route": {"type": "choice", "criteria": {...}}})
```
The Python host builds prompts with upstream's own runtime (`d3_runtime.py` in the upstream snapshot, run without weights), so prompts match upstream exactly.

License: Apache-2.0, as upstream.