Salastil/odysseus
1
0
Fork 0
mirror of https://github.com/pewdiepie-archdaemon/odysseus.git synced 2026-06-17 10:15:27 -04:00
Code Issues Packages Projects Releases Wiki Activity

Fix native Cookbook quant classification

Browse Source
This commit is contained in:
spooky
2026-06-02 14:07:20 +10:00
committed by GitHub
parent 65b5d65059
commit cd4f496cb4
6 changed files with 201 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files
services/hwfit/fit.py
+30 -27
View File
@@ -219,9 +219,9 @@ def _quant_bits(q):
Returns 0 when unknown (caller treats unknown as "don't filter")."""
qu = (q or "").upper().replace("-", "").replace("_", "").replace(" ", "")
# GGUF k-quants + float formats
if qu.startswith("Q8") or "FP8" in qu:
if qu.startswith("Q8") or "FP8" in qu or "INT8" in qu or qu.startswith("W8"):
return 8
if qu.startswith("Q4") or qu.startswith("IQ4"):
if qu.startswith("Q4") or qu.startswith("IQ4") or "FP4" in qu or "NF4" in qu or "INT4" in qu or qu.startswith("W4"):
return 4
if qu.startswith("Q2") or qu.startswith("IQ2"):
return 2
@@ -233,7 +233,7 @@ def _quant_bits(q):
return 6
if qu.startswith("F16") or qu.startswith("BF16") or qu.startswith("F32"):
return 16
# Prequantized formats: pull the bit-width digit (AWQ4 / AWQ4BIT / GPTQ8 / 4BIT / INT8 …)
# Prequantized formats: pull the bit-width digit (AWQ4 / AWQ4BIT / GPTQ8 / 4BIT / INT8 ...)
m = re.search(r"(?:AWQ|GPTQ|MLX|EXL2|BNB|INT|W)(\d{1,2})", qu) or re.search(r"(\d{1,2})BIT", qu)
if m:
b = int(m.group(1))
@@ -282,15 +282,21 @@ def analyze_model(model, system, target_quant=None, scoring_use_case=None):
else:
effective_vram = gpu_vram
native_gpu_only = preq and not native_quant.startswith("mlx-")
# Determine which quant to evaluate at
native_quant_prefixes = (
"AWQ-", "GPTQ-", "FP8", "FP4", "NVFP4", "MXFP4", "NF4",
"INT4", "INT8", "W4A16", "W8A8", "W8A16",
)
if preq:
# AWQ/GPTQ/FP8/MLX come at a fixed bit-width. If the user picked a
# GGUF quant tier (Q4/Q8/etc.), do not treat a same-bit AWQ/GPTQ build
# as equivalent. "Q4" means llama.cpp/Ollama-style GGUF in this UI;
# AWQ/GPTQ/FP8 are separate GPU-serving formats and must only appear
# when explicitly selected or when no quant filter is applied.
# Native HF/vLLM quantized repos come at a fixed format. If the user
# picked a GGUF quant tier (Q4/Q8/etc.), do not treat same-bit
# AWQ/GPTQ/FP8/FP4 builds as equivalent; those formats are separate
# serving paths and only appear when explicitly selected or unfiltered.
if target_quant:
if not any(target_quant.startswith(p) for p in ("AWQ-", "GPTQ-", "FP8", "NVFP4")):
if not any(target_quant.startswith(p) for p in native_quant_prefixes):
return None
_tb, _nb = _quant_bits(target_quant), _quant_bits(native_quant)
if _tb and _nb and _tb != _nb:
@@ -303,16 +309,7 @@ def analyze_model(model, system, target_quant=None, scoring_use_case=None):
# Default: Q4_K_M (user's stated preference)
quant_to_try = "Q4_K_M"
result = _try_quant_at(model, quant_to_try, ctx, effective_vram, eff_ram)
# If target quant doesn't fit and it's not pre-quantized, try lower quants
if result is None and not preq and target_quant:
from services.hwfit.models import QUANT_HIERARCHY
idx = QUANT_HIERARCHY.index(target_quant) if target_quant in QUANT_HIERARCHY else -1
for q in QUANT_HIERARCHY[idx + 1:]:
result = _try_quant_at(model, q, ctx, effective_vram, eff_ram)
if result:
break
result = _try_quant_at(model, quant_to_try, ctx, effective_vram, 0 if native_gpu_only else eff_ram)
if result is None:
# Model doesn't fit on the user's current hardware. Surface it
@@ -447,8 +444,11 @@ def rank_models(system, use_case=None, limit=50, search=None, sort="score", quan
results.sort(key=sort_fn, reverse=(sort != "vram"))
return results[:limit]
# If user picked a prequantized format (AWQ/FP8/GPTQ/NVFP4), filter to only those models
filter_native = quant and any(quant.startswith(p) for p in ("AWQ-", "GPTQ-", "FP8", "NVFP4"))
# If user picked a native prequantized format, filter to only those models.
filter_native = quant and any(quant.startswith(p) for p in (
"AWQ-", "GPTQ-", "FP8", "FP4", "NVFP4", "MXFP4", "NF4",
"INT4", "INT8", "W4A16", "W8A8", "W8A16",
))
system_backend = (system.get("backend") or "").lower()
apple_silicon = system_backend in ("mps", "metal", "apple")
@@ -459,9 +459,9 @@ def rank_models(system, use_case=None, limit=50, search=None, sort="score", quan
if "nvfp4" in (m.get("name") or "").lower():
native_q = "NVFP4"
# MLX is Apple Silicon only. Hide MLX rows on non-Mac hardware scans,
# but leave them visible on Metal/MPS so Mac support is not broken.
if not apple_silicon and (native_q.startswith("mlx-") or "mlx" in (m.get("name") or "").lower()):
# MLX needs the mlx_lm runtime, which Odysseus does not generate serve
# commands for. Hide it on every backend, including Metal.
if native_q.startswith("mlx-") or "mlx" in (m.get("name") or "").lower():
continue
# ROCm support for vLLM/SGLang quantized safetensors is too brittle to
@@ -479,20 +479,23 @@ def rank_models(system, use_case=None, limit=50, search=None, sort="score", quan
# default GGUF quant) and vLLM-only AWQ/GPTQ/FP8 builds alike. Without
# this the Cookbook recommends models the Mac can't run; on CUDA these
# stay visible because vLLM serves safetensors directly.
is_mlx = native_q.startswith("mlx-") or "mlx" in (m.get("name") or "").lower()
if apple_silicon and not (m.get("is_gguf") or m.get("gguf_sources") or is_mlx):
if apple_silicon and not (m.get("is_gguf") or m.get("gguf_sources")):
continue
# Format filter: AWQ tab → only AWQ models, FP8 tab → only FP8 models
# Format filter: AWQ tab -> only AWQ models, FP4 tab -> FP4-family models, etc.
if filter_native:
if quant == "FP8" and native_q != "FP8":
continue
if quant == "FP4" and native_q not in ("FP4", "NVFP4", "MXFP4", "NF4"):
continue
if quant.startswith("AWQ") and not native_q.startswith("AWQ"):
continue
if quant.startswith("GPTQ") and not native_q.startswith("GPTQ"):
continue
if quant.startswith("NVFP4") and not native_q.startswith("NVFP4"):
continue
if quant in ("INT4", "INT8", "W4A16", "W8A8", "W8A16") and native_q != quant:
continue
if search:
name = m.get("name", "").lower()