Clip Vit B32 Litert

Files

• clip_image_encoder.tflite — image encoder, NCHW [1, 3, 224, 224] → [1, 512] (L2-normalized).
• text_embeddings.bin — pre-computed text embeddings for 96 labels ([96, 512], prompt "a photo of a {label}"). Little-endian: int32 num_labels, int32 dim, float32[num_labels*dim].
• labels.txt — the 96 labels, one per line.

Preprocessing / use

RGB → center-crop to square → resize 224×224 → CLIP normalization (mean = [0.4815, 0.4578, 0.4082] × 255, std = [0.2686, 0.2613, 0.2758] × 255), planar NCHW. The output is L2-normalized; score labels by cosine similarity against text_embeddings.bin, then softmax with logit scale 100.

GPU-compatibility note

A stock conversion does not run on the ML Drift GPU delegate: torch.nn.MultiheadAttention lowers to 5D RESHAPE tensors (GPU max is 4D), so the model fails to compile. This export uses a 4D manual-attention rewrite (nn.MultiheadAttention → explicit 4D matmul + softmax, weights copied verbatim → numerically exact) plus the standard GELU → x·sigmoid(1.702x) approximation. With that, the whole encoder is GPU-clean (691/691 ops on the delegate).

License

MIT (OpenAI CLIP / OpenCLIP).

Want a different model on-device? Open a request — free, open weights only; the export and its measured numbers get published publicly.

--- license: mit library_name: litert pipeline_tag: zero-shot-image-classification tags: - clip - zero-shot-image-classification - litert - tflite - on-device - gpu --- # CLIP ViT-B/32 image encoder — LiteRT (TFLite) GPU OpenAI / OpenCLIP **CLIP ViT-B/32** image encoder converted with **litert-torch** for **LiteRT `CompiledModel` GPU (ML Drift)** on Android. Verified end-to-end on a Pixel 8a: the full graph (**691/691 ops**) runs on the OpenCL GPU delegate at **~40 ms/inference**. ## Files - `clip_image_encoder.tflite` — image encoder, NCHW `[1, 3, 224, 224]` → `[1, 512]` (L2-normalized). - `text_embeddings.bin` — pre-computed text embeddings for 96 labels (`[96, 512]`, prompt `"a photo of a {label}"`). Little-endian: `int32 num_labels, int32 dim, float32[num_labels*dim]`. - `labels.txt` — the 96 labels, one per line. ## Preprocessing / use RGB → center-crop to square → resize 224×224 → CLIP normalization (`mean = [0.4815, 0.4578, 0.4082] × 255`, `std = [0.2686, 0.2613, 0.2758] × 255`), planar NCHW. The output is L2-normalized; score labels by cosine similarity against `text_embeddings.bin`, then softmax with logit scale 100. ## GPU-compatibility note A stock conversion does **not** run on the ML Drift GPU delegate: `torch.nn.MultiheadAttention` lowers to **5D RESHAPE** tensors (GPU max is 4D), so the model fails to compile. This export uses a **4D manual-attention rewrite** (`nn.MultiheadAttention` → explicit 4D matmul + softmax, weights copied verbatim → numerically exact) plus the standard `GELU → x·sigmoid(1.702x)` approximation. With that, the whole encoder is GPU-clean (691/691 ops on the delegate). ## License MIT (OpenAI CLIP / OpenCLIP). <!-- funnel:v1 --> --- **Want a different model on-device?** [Open a request](https://github.com/john-rocky/on-device-requests) — free, open weights only; the export and its measured numbers get published publicly. <!-- /funnel:v1 -->