Merge remote-tracking branch 'origin/main' into visual-pr-playground

# Conflicts:
#	routes/cookbook_routes.py
#	routes/hwfit_routes.py
#	services/hwfit/fit.py
#	services/hwfit/models.py
#	static/js/cookbook-diagnosis.js
#	static/js/cookbook-hwfit.js
#	static/js/cookbook.js
#	static/js/cookbookRunning.js

services/hwfit/models.py

@@ -5,7 +5,9 @@ import re
 QUANT_HIERARCHY = ["Q8_0", "Q6_K", "Q5_K_M", "Q4_K_M", "Q3_K_M", "Q2_K"]
 
 QUANT_BPP = {
-    "F32": 4.0, "F16": 2.0, "BF16": 2.0, "FP8": 1.0, "INT8": 1.0, "NVFP4": 0.5,
+    "F32": 4.0, "F16": 2.0, "BF16": 2.0, "FP8": 1.0,
+    "FP4": 0.50, "NVFP4": 0.50, "MXFP4": 0.50, "NF4": 0.50,
+    "INT4": 0.50, "INT8": 1.0, "W4A16": 0.50, "W8A8": 1.0, "W8A16": 1.0,
     "Q8_0": 1.05, "Q6_K": 0.80, "Q5_K_M": 0.68,
     "Q4_K_M": 0.58, "Q4_0": 0.58, "Q3_K_M": 0.48, "Q2_K": 0.37,
     "AWQ-4bit": 0.50, "AWQ-8bit": 1.0,
@@ -14,7 +16,9 @@ QUANT_BPP = {
 }
 
 QUANT_SPEED_MULT = {
-    "F16": 0.6, "BF16": 0.6, "FP8": 0.85, "INT8": 0.85, "NVFP4": 1.1,
+    "F16": 0.6, "BF16": 0.6, "FP8": 0.85,
+    "FP4": 1.15, "NVFP4": 1.15, "MXFP4": 1.15, "NF4": 1.10,
+    "INT4": 1.15, "INT8": 0.85, "W4A16": 1.15, "W8A8": 0.85, "W8A16": 0.85,
     "Q8_0": 0.8, "Q6_K": 0.95, "Q5_K_M": 1.0,
     "Q4_K_M": 1.15, "Q4_0": 1.15, "Q3_K_M": 1.25, "Q2_K": 1.35,
     "AWQ-4bit": 1.2, "AWQ-8bit": 0.85,
@@ -23,8 +27,10 @@ QUANT_SPEED_MULT = {
 }
 
 QUANT_QUALITY_PENALTY = {
-    "F16": 0.0, "BF16": 0.0, "FP8": 0.0, "INT8": 0.0, "NVFP4": -0.5,
-    "Q8_0": -0.5, "Q6_K": -1.5, "Q5_K_M": -2.5,
+    "F16": 0.0, "BF16": 0.0, "FP8": 0.0,
+    "FP4": -3.0, "NVFP4": -3.0, "MXFP4": -3.0, "NF4": -4.0,
+    "INT4": -4.0, "INT8": 0.0, "W4A16": -4.0, "W8A8": 0.0, "W8A16": 0.0,
+    "Q8_0": 0.0, "Q6_K": -1.0, "Q5_K_M": -2.0,
     "Q4_K_M": -5.0, "Q4_0": -5.0, "Q3_K_M": -8.0, "Q2_K": -12.0,
     # Bare "AWQ" and "AWQ-8bit" used to be 0.0 (tied with FP8). In practice
     # AWQ-anything is a calibrated reconstruction, not raw 8-bit weights —
@@ -36,7 +42,9 @@ QUANT_QUALITY_PENALTY = {
 }
 
 QUANT_BYTES_PER_PARAM = {
-    "F16": 2.0, "BF16": 2.0, "FP8": 1.0, "INT8": 1.0, "NVFP4": 0.5,
+    "F16": 2.0, "BF16": 2.0, "FP8": 1.0,
+    "FP4": 0.5, "NVFP4": 0.5, "MXFP4": 0.5, "NF4": 0.5,
+    "INT4": 0.5, "INT8": 1.0, "W4A16": 0.5, "W8A8": 1.0, "W8A16": 1.0,
     "Q8_0": 1.0, "Q6_K": 0.75, "Q5_K_M": 0.625,
     "Q4_K_M": 0.5, "Q4_0": 0.5, "Q3_K_M": 0.375, "Q2_K": 0.25,
     "AWQ-4bit": 0.5, "AWQ-8bit": 1.0,
@@ -44,8 +52,55 @@ QUANT_BYTES_PER_PARAM = {
     "mlx-4bit": 0.5, "mlx-8bit": 1.0, "mlx-6bit": 0.75,
 }
 
-# Pre-quantized formats that should NOT go through the GGUF quant hierarchy
-PREQUANTIZED_PREFIXES = ("AWQ-", "GPTQ-", "mlx-", "FP8", "INT8", "NVFP4")
+# Pre-quantized formats that should NOT go through the GGUF quant hierarchy.
+# These are native HF/vLLM-style repos, not llama.cpp GGUF quant tiers.
+PREQUANTIZED_PREFIXES = (
+    "AWQ-", "GPTQ-", "mlx-", "FP8", "FP4", "NVFP4", "MXFP4", "NF4",
+    "INT4", "INT8", "W4A16", "W8A8", "W8A16",
+)
+
+
+def infer_quantization_from_name(name):
+    n = (name or "").lower()
+    if "nvfp4" in n:
+        return "NVFP4"
+    if "mxfp4" in n:
+        return "MXFP4"
+    if re.search(r"(^|[-_/])nf4($|[-_/])", n):
+        return "NF4"
+    if re.search(r"(^|[-_/])fp4($|[-_/])", n):
+        return "FP4"
+    if re.search(r"(^|[-_/])w4a16($|[-_/])", n):
+        return "W4A16"
+    if re.search(r"(^|[-_/])w8a8($|[-_/])", n):
+        return "W8A8"
+    if re.search(r"(^|[-_/])w8a16($|[-_/])", n):
+        return "W8A16"
+    is8 = "8bit" in n or "8-bit" in n or "int8" in n
+    if "awq" in n:
+        return "AWQ-8bit" if is8 else "AWQ-4bit"
+    if "gptq" in n:
+        return "GPTQ-Int8" if is8 else "GPTQ-Int4"
+    if "mlx" in n:
+        if "6bit" in n:
+            return "mlx-6bit"
+        return "mlx-8bit" if is8 else "mlx-4bit"
+    if "fp8" in n:
+        return "FP8"
+    if "int4" in n or "4bit" in n or "4-bit" in n:
+        return "INT4"
+    if "int8" in n or "8bit" in n or "8-bit" in n:
+        return "INT8"
+    return ""
+
+
+def _normalize_model_entry(model):
+    if not isinstance(model, dict):
+        return model
+    inferred = infer_quantization_from_name(model.get("name", ""))
+    if inferred and (model.get("quantization") in (None, "", "Q4_K_M") or model.get("_discovered")):
+        model["quantization"] = inferred
+    return model
 
 
 def is_prequantized(model):
@@ -72,7 +127,13 @@ def params_b(model):
         pc = pc.strip().upper()
         m = re.match(r"^([\d.]+)\s*([BKMGT]?)$", pc)
         if m:
-            val = float(m.group(1))
+            try:
+                val = float(m.group(1))
+            except ValueError:
+                # Malformed count like "1.5.3B" — [\d.]+ matches but float()
+                # rejects it. One bad catalog row must not abort the whole
+                # ranking pass, so treat it as unknown size.
+                return 0.0
             suffix = m.group(2)
             if suffix == "B":
                 return val
@@ -180,7 +241,7 @@ def get_models():
         data_path = os.path.join(os.path.dirname(__file__), "data", "hf_models.json")
         try:
             with open(data_path, encoding="utf-8") as f:
-                _models_cache = json.load(f)
+                _models_cache = [_normalize_model_entry(m) for m in json.load(f)]
         except (FileNotFoundError, json.JSONDecodeError):
             _models_cache = []
     return _models_cache