Upload README.md with huggingface_hub

README.md

@@ -0,0 +1,61 @@
+---
+license: apache-2.0
+base_model: Qwen/Qwen3.5-0.8B
+tags:
+  - mlx
+  - qwen3.5
+  - text-generation
+  - on-device
+  - apple-silicon
+  - int4
+  - int8
+language:
+  - en
+  - zh
+  - ja
+  - ko
+  - de
+  - es
+  - fr
+library_name: mlx
+pipeline_tag: text-generation
+---
+
+# Qwen3.5-0.8B Chat — MLX (Apple Silicon)
+
+Text-only extraction of [Qwen3.5-0.8B](https://huggingface.co/Qwen/Qwen3.5-0.8B) for on-device LLM chat on Apple Silicon via [MLX](https://github.com/ml-explore/mlx).
+
+## Architecture
+
+Qwen3.5 is a **hybrid** model with 24 layers:
+- **18× DeltaNet** — linear attention with gated delta rule recurrence, O(1) memory per step
+- **6× GatedAttention** — full scaled dot-product attention with KV cache, partial RoPE (25%)
+- Pattern: `[linear, linear, linear, full] × 6`
+- Tied word embeddings (lm_head = embed_tokens)
+
+## Variants
+
+| Variant | Size | Path |
+|---------|------|------|
+| INT4 | 404 MB | `int4/model.safetensors` |
+| INT8 | 763 MB | `int8/model.safetensors` |
+
+Each variant includes `config.json`, `tokenizer.json`, and `tokenizer_config.json`.
+
+## Usage
+
+```swift
+import Qwen3Chat
+
+let model = try await Qwen35MLXChat.fromPretrained(quantization: .int4)
+let response = try model.generate(
+    messages: [ChatMessage(role: .user, content: "Hello!")],
+    sampling: ChatSamplingConfig(temperature: 0.3, maxTokens: 100)
+)
+```
+
+Part of the [soniqo](https://soniqo.audio) speech toolkit for Apple Silicon.
+
+## Source
+
+Repackaged from [mlx-community/Qwen3.5-0.8B-4bit](https://huggingface.co/mlx-community/Qwen3.5-0.8B-4bit) and [mlx-community/Qwen3.5-0.8B-MLX-8bit](https://huggingface.co/mlx-community/Qwen3.5-0.8B-MLX-8bit) — vision tower removed, text model only.