Update README.md to include RabbitLLM logo, clarify compatibility with Qwen2 and Qwen3 models, and provide detailed architecture support status. Remove outdated macOS installation instructions and enhance model loading examples for better user guidance.

Author: ManuelSLemos

README.md

@@ -1,15 +1,21 @@
 # RabbitLLM
 
+![RabbitLLM logo](assets/logo-rabbitllm.jpg)
+
 **Run 70B+ LLMs on a single 4GB GPU — no quantization required.**
 
 [![PyPI](https://img.shields.io/pypi/v/rabbitllm?color=blue)](https://pypi.org/project/rabbitllm/)
 [![Python 3.10+](https://img.shields.io/badge/python-3.10%2B-blue)](https://www.python.org/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-green.svg)](LICENSE)
 [![CI](https://github.com/manuelslemos/rabbitllm/actions/workflows/ci.yml/badge.svg)](https://github.com/manuelslemos/rabbitllm/actions)
 
-RabbitLLM enables inference on large language models (70B+ parameters) on consumer GPUs with as
-little as 4GB VRAM by streaming model layers one at a time through GPU memory. No quantization,
-distillation, or pruning needed — full model quality.
+RabbitLLM is a **fork of [AirLLM](https://github.com/airllm/airllm)**. It enables inference on large language models (70B+ parameters) on consumer GPUs with as little as 4GB VRAM by streaming model layers one at a time through GPU memory. No quantization, distillation, or pruning needed — full model quality.
+
+### Compatibility (current status)
+
+- **Tested and supported:** only **Qwen2** and **Qwen3** are currently tested and compatible. Use these families for reliable results.
+- **Other architectures** (Llama, Mistral, Mixtral, etc.) are present in the codebase but **not yet compatible** — use at your own risk.
+- **Apple (macOS / Apple Silicon)** is **not supported**; run on Linux or Windows with a CUDA-capable GPU (or CPU fallback on x86/ARM Linux).
 
 ## How it works
 
@@ -42,7 +48,7 @@ If the prebuilt wheel is unavailable for your setup, install from
 ```python
 from rabbitllm import AutoModel
 
-model = AutoModel.from_pretrained("meta-llama/Llama-3-8B")
+model = AutoModel.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct")  # or any Qwen2 / Qwen3
 
 input_tokens = model.tokenizer(
     ["What is the capital of France?"],
@@ -68,19 +74,21 @@ no need to pick the right class manually.
 
 ## Supported models
 
-| Family | Architectures | Class |
-|---|---|---|
-| Llama 2 / 3 / 3.1 / 3.2 | `LlamaForCausalLM` | `RabbitLLMLlama2` |
-| Qwen2 / Qwen2.5 / Qwen3 | `Qwen2ForCausalLM`, `Qwen3ForCausalLM` | `RabbitLLMQWen2` |
-| Qwen v1 | `QWenLMHeadModel` | `RabbitLLMQWen` |
-| Mistral | `MistralForCausalLM` | `RabbitLLMMistral` |
-| Mixtral | `MixtralForCausalLM` | `RabbitLLMMixtral` |
-| InternLM | `InternLMForCausalLM` | `RabbitLLMInternLM` |
-| ChatGLM | `ChatGLMModel` | `RabbitLLMChatGLM` |
-| Baichuan | `BaichuanForCausalLM` | `RabbitLLMBaichuan` |
-| Gemma 2 / 3 | `Gemma2ForCausalLM`, `Gemma3ForCausalLM` | `RabbitLLMLlama2` |
-| DeepSeek V2 / V3 | `DeepseekV2ForCausalLM`, `DeepseekV3ForCausalLM` | `RabbitLLMLlama2` |
-| Phi 2 / 3 / 4 | `Phi3ForCausalLM`, `Phi4ForCausalLM` | `RabbitLLMLlama2` |
+**Only Qwen2 and Qwen3 are tested and supported.** The following table lists the architectures present in the codebase; others are not yet compatible.
+
+| Family | Architectures | Class | Status |
+|---|---|---|---|
+| **Qwen2 / Qwen2.5 / Qwen3** | `Qwen2ForCausalLM`, `Qwen3ForCausalLM` | `RabbitLLMQWen2` | **Tested, supported** |
+| Llama 2 / 3 / 3.1 / 3.2 | `LlamaForCausalLM` | `RabbitLLMLlama2` | Not yet compatible |
+| Qwen v1 | `QWenLMHeadModel` | `RabbitLLMQWen` | Not yet compatible |
+| Mistral | `MistralForCausalLM` | `RabbitLLMMistral` | Not yet compatible |
+| Mixtral | `MixtralForCausalLM` | `RabbitLLMMixtral` | Not yet compatible |
+| InternLM | `InternLMForCausalLM` | `RabbitLLMInternLM` | Not yet compatible |
+| ChatGLM | `ChatGLMModel` | `RabbitLLMChatGLM` | Not yet compatible |
+| Baichuan | `BaichuanForCausalLM` | `RabbitLLMBaichuan` | Not yet compatible |
+| Gemma 2 / 3 | `Gemma2ForCausalLM`, `Gemma3ForCausalLM` | `RabbitLLMLlama2` | Not yet compatible |
+| DeepSeek V2 / V3 | `DeepseekV2ForCausalLM`, `DeepseekV3ForCausalLM` | `RabbitLLMLlama2` | Not yet compatible |
+| Phi 2 / 3 / 4 | `Phi3ForCausalLM`, `Phi4ForCausalLM` | `RabbitLLMLlama2` | Not yet compatible |
 
 Unknown architectures fall back to the Llama-based implementation with a warning.
 
@@ -109,7 +117,7 @@ Block-wise quantization reduces on-disk and in-memory layer size:
 - **8-bit**: ~50% of original size.
 
 ```python
-model = AutoModel.from_pretrained("mistralai/Mistral-7B-v0.1", compression="4bit")
+model = AutoModel.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct", compression="4bit")
 ```
 
 Requires `bitsandbytes`: `pip install bitsandbytes`.
@@ -119,7 +127,7 @@ Requires `bitsandbytes`: `pip install bitsandbytes`.
 Pass a HuggingFace token for repos that require access approval:
 
 ```python
-model = AutoModel.from_pretrained("meta-llama/Llama-2-7b-hf", token="hf_YOUR_TOKEN")
+model = AutoModel.from_pretrained("Qwen/Qwen2.5-7B-Instruct", token="hf_YOUR_TOKEN")
 ```
 
 Or set the `HF_TOKEN` environment variable.
@@ -135,16 +143,6 @@ export HF_HOME="$(pwd)/models"
 The `models/` directory is in `.gitignore`. RabbitLLM will store split layers alongside
 the HuggingFace cache.
 
-## macOS / Apple Silicon
-
-Install and run the same way. Requires [mlx](https://github.com/ml-explore/mlx) and PyTorch.
-Only Apple Silicon is supported.
-
-```bash
-pip install mlx torch
-python -c "from rabbitllm import AutoModel; model = AutoModel.from_pretrained('TinyLlama/TinyLlama-1.1B-Chat-v1.0')"
-```
-
 ## Documentation
 
 - [docs/ARCHITECTURE.md](docs/ARCHITECTURE.md) — Design decisions: layer-streaming, KV cache, tied weights, attention implementations.