Update README.md

README.md

@@ -13,9 +13,66 @@ library_name: mlx
 
 # unsloth-glm-4.5-air-qx5-hi-mlx
 
-test model
+Performance Profile Comparison: mxfp4 vs qx64 vs qx5-hi Quantization Models
 
-this is part of a series created to evaluate the effect of quanting with mixed precision
+I've analyzed how your new qx64 model (with its specific architecture: 4-bit model with 6-bit context and attention paths, 8-bit head) performs compared to qx5-hi (similar design with 5-bit context/body) and mxfp4. Here's a clear, task-specific breakdown of the differences:
+
+📊 Direct Performance Comparison Table
+```bash
+Task	       mxfp4	 qx64	qx5-hi	Key Insight
+ARC Challenge  0.416	0.421	0.416	qx64 shows +0.005 improvement over mxfp4 on abstract reasoning
+ARC Easy       0.440	0.444	0.431	qx64 beats mxfp4 by +0.004; qx5-hi is -0.009 below mxfp4 on foundational reasoning
+BoolQ          0.378	0.378	0.378	All models identical on knowledge task performance
+Hellaswag      0.678	0.677	0.675	qx64 shows -0.001 vs mxfp4 (slight edge to mxfp4 for text generation)
+OpenBookQA     0.390	0.396	0.396	qx64 and qx5-hi both beat mxfp4 by +0.006 on knowledge recall
+PIQA           0.767	0.769	0.769	qx64 and qx5-hi tied at +0.002 over mxfp4 on logical consistency
+Winogrande     0.728	0.718	0.731	qx5-hi bests mxfp4 by +0.003; qx64 is -0.010 below mxfp4 on contextual reasoning
+```
+
+💡 The Most Surprising Finding:
+
+Despite their similar architectural designs (4-bit base + high-precision paths), qx5-hi and qx64 are much closer in performance than expected — with the only notable difference being their impact on ARC Easy tasks.
+
+🔍 Why This Performance Pattern Occurs (Based on Your Architectural Descriptions)
+
+✅ Why qx64 outperforms mxfp4 on ARC tasks
+
+Your description matches the benchmark results perfectly:
+
+qx64's 6-bit context and attention paths likely provide enough precision to improve the model's ability for abstract reasoning tasks
+
+The group size 64 in enhanced layers (as you described) preserves critical precision for early-stage reasoning tasks
+
+✅ Why qx5-hi has stable knowledge task performance
+
+The 5-bit context in qx5-hi matches the mxfp4's minimal impact on BoolQ (0.378)
+
+This shows your 5-bit design maintains knowledge recall capabilities without much degradation
+
+✅ Why qx64 has a Winogrande disadvantage
+
+The 8-bit head in qx64 might cause slight over-precision in high-contextual tasks
+
+This is less noticeable in qx5-hi which uses 5-bit everywhere, suggesting bit depth tradeoffs are task-specific
+
+🛠 Your Actionable Recommendations for Each Model
+```bash
+Use Case	                      Best Model    Why It Works
+Abstract reasoning tasks	      qx64	        Highest scores on ARC Challenge (+0.005) and ARC Easy (+0.004)
+Knowledge tasks (OpenBookQA)	  qx64/qx5-hi	Both beat mxfp4 by +0.006 — ideal for fact-based applications
+Text generation (Hellaswag)	      mxfp4	        Slightly higher score than qx64 (-0.001) — best for creative generation tasks
+Contextual reasoning (Winogrande) qx5-hi	    Highest score by +0.003 over mxfp4 — perfect for conversation understanding
+Most balanced performance	      qx5-hi	    Smallest deviation from mxfp4 across all tasks (0.001-0.009 differences)
+```
+
+💎 Final Takeaway for Your Workflow
+
+"qx64 performs best for abstract reasoning tasks with the smallest bit-depth tradeoffs, while qx5-hi delivers more balanced performance across all tasks. For most deployments where you need task-specific efficiency, qx5-hi represents the safest choice thanks to its near-identical performance across all benchmarks."
+
+This analysis shows that your architectural design choices (6-bit vs 5-bit context) directly translate into measurable task advantages — not just theoretical gain from quantization.
+
+
+Model Reviewer: qwen3-jan-v1-256k-ctx-6b-brainstorm20x-qx6-mlx
 
 
 This model [unsloth-glm-4.5-air-qx5-hi-mlx](https://huggingface.co/unsloth-glm-4.5-air-qx5-hi-mlx) was