Merge branch 'main' of https://github.com/ashvardanian/uform

.gitignore

@@ -4,7 +4,13 @@ test
 build/
 package-lock.json
 *.egg-info
-*.onnx
 __pycache__
 .build
-.swiftpm
+.swiftpm
+.hf_token
+node_modules
+
+# Tensors & ML Model
+*.onnx
+*.pt
+*.safetensors

README.md

@@ -20,9 +20,11 @@ For Content Understanding and Generation<br/>
 <p align="center">
 Multimodal Embeddings from 64 to 768 Dimensions • 1B Parameter Chat
 <br/>
-Short Texts • Images • 🔜 Video Clips
+Short Texts • Images • 🔜 Video Clips • 🔜 Long Documents
 <br/>
-PyTorch • ONNX
+ONNX • CoreML • PyTorch
+<br/>
+Python • JavaScript • Swift
 </p>
 
 ---
@@ -279,7 +281,7 @@ The generative model can be used to caption images, summarize their content, or
 The exact behavior is controlled by prompts.
 
 ```python
-from uform.gen_model import VLMForCausalLM, VLMProcessor
+from uform.torch_decoders import VLMForCausalLM, VLMProcessor
 
 model = VLMForCausalLM.from_pretrained('unum-cloud/uform-gen')
 processor = VLMProcessor.from_pretrained('unum-cloud/uform-gen')

javascript/README.md

@@ -0,0 +1,10 @@
+# UForm for JavaScript
+
+
+
+```bash
+pnpm add uform
+npm add uform
+yarn add uform
+```
+

package.json

@@ -0,0 +1,11 @@
+{
+    "name": "uform",
+    "private": true,
+    "version": "2.0.2",
+    "description": "Pocket-Sized Multimodal AI for Content Understanding and Generation",
+    "dependencies": {
+        "@huggingface/hub": "^0.14.8",
+        "@xenova/transformers": "^2.17.0",
+        "onnxruntime-web": "^1.17.3"
+    }
+}

python/scripts/bench.py

@@ -13,7 +13,7 @@
 )
 
 from uform import get_model
-from uform.gen_model import VLMForCausalLM, VLMProcessor
+from uform.torch_decoders import VLMForCausalLM, VLMProcessor
 
 dtype = torch.bfloat16
 low_cpu_mem_usage = False