Readme tweaks (#5)

- Readme tweaks (57642a0f98520fdf8891310b3adfbcbb9042e79a)

README.md

@@ -69,16 +69,14 @@ The MoE router gates (`mlp.gate` and `mlp.shared_expert_gate` for all 60 layers)
 
 ## Requirements
 
-- Apple Silicon Mac with **at least 256 GB unified memory** (e.g., Mac Studio M2/M3/M4 Ultra 256GB+)
+- Apple Silicon Mac with **at least 256 GB unified memory** (e.g., Mac Studio M3 Ultra 256GB+)
 - Python 3.10+
-- [`mlx-lm`](https://github.com/ml-explore/mlx-lm) v0.30.7 or better
-
-> **Note**: Although only ~17B parameters are active per token, all 397B parameters (~223 GB quantized) must be loaded into unified memory.
+- [`mlx-lm`](https://github.com/ml-explore/mlx-lm) from the `main` branch
 
 ## Installation
 
 ```bash
-pip install mlx-lm
+pip install git+https://github.com/ml-explore/mlx-lm
 ```
 
 ## Usage