ltcai 5.1.0 → 5.3.0

package/latticeai/services/model_capability_registry.py

@@ -0,0 +1,482 @@
+"""Structured Model Capability Registry for Lattice AI 5.2.0+.
+
+User-focused, transparent model catalog with:
+- HF repo provenance
+- Modality / vision support
+- Quantization, size, download/load strategies
+- Hardware notes (RAM estimates, Apple Silicon affinity)
+- License / safety notes
+- Verification status (populated by scripts/verify_hf_model_registry.py or runtime light probe)
+
+This replaces the flat ENGINE_MODEL_CATALOG construction with a richer,
+queryable source of truth while preserving exact legacy shapes for
+model_catalog / recommendation / API / frontend consumers.
+
+All entries are recommended multimodal (VLM) first. Text-only can be added later.
+Verification is honest: hf_exists + light metadata/config presence; full weights
+are never auto-fetched by the verifier. Large models (>12GB) explicitly note
+"local load practical only on high-RAM Apple Silicon or CUDA; expect long download".
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass, field
+from typing import Any, Dict, List, Optional
+
+
+@dataclass(frozen=True)
+class HardwareProfile:
+    min_ram_gb: Optional[float] = None
+    recommended_ram_gb: Optional[float] = None
+    apple_silicon_pref: bool = False
+    cuda_pref: bool = False
+    notes: str = ""
+
+
+@dataclass(frozen=True)
+class VerificationStatus:
+    hf_exists: bool = False
+    hf_last_checked: Optional[str] = None  # ISO
+    has_config: bool = False
+    has_tokenizer: bool = False
+    has_weights_hint: bool = False  # safetensors or gguf siblings seen in meta
+    pipeline_tag: Optional[str] = None
+    likes: Optional[int] = None
+    license: Optional[str] = None
+    notes: str = ""
+    verified_by: str = "hf-api-light"  # or "local-load-test"
+
+
+@dataclass(frozen=True)
+class ModelCapability:
+    """Rich capability entry. id is the canonical key used in ENGINE_MODEL_CATALOG."""
+    id: str
+    hf_repo_id: str  # clean HF path for download (e.g. mlx-community/xxx or Qwen/yyy)
+    name: str
+    family: str
+    tag: str
+    size: str  # display string "7.6GB", "pull required"
+    modality: str = "multimodal"  # multimodal | vision | text | audio etc.
+    quantization: Optional[str] = None  # "4bit", "Q4_K_M", "GGUF-Q4"
+    provider_hints: List[str] = field(default_factory=lambda: ["local_mlx"])  # which engines this id primarily maps to
+    download_strategy: str = "hf_hub"  # hf_hub | ollama_pull | lmstudio_app | gguf_manual
+    load_strategy: str = "mlx_vlm"  # mlx_vlm | ollama | vllm | lmstudio | llamacpp_server
+    hardware: HardwareProfile = field(default_factory=HardwareProfile)
+    license: str = "apache-2.0"
+    safety_notes: str = "Standard open weights; review license and responsible use guidelines."
+    source_country: str = ""
+    source_company: str = ""
+    execution_method: str = "내 컴퓨터에서만 실행"
+    internet_requirement: str = "모델을 다운로드할 때만 인터넷 필요; 실행 중에는 필요 없음"
+    # Verification
+    verification: VerificationStatus = field(default_factory=VerificationStatus)
+    # UI / rec hints
+    recommended_default: bool = False
+    display_priority: int = 100
+
+    def to_legacy_dict(self) -> Dict[str, Any]:
+        """Exact shape expected by older ENGINE_MODEL_CATALOG consumers + extra 5.2 fields."""
+        base = {
+            "id": self.id,
+            "name": self.name,
+            "model_name": self.name.split(" via ")[0] if " via " in self.name else self.name,
+            "family": self.family,
+            "tag": self.tag,
+            "size": self.size,
+            "pullable": True,
+            "modality": self.modality,
+            "source_country": self.source_country,
+            "source_company": self.source_company,
+            "execution_method": self.execution_method,
+            "run_location": "내 컴퓨터에서만 실행",
+            "internet_requirement": self.internet_requirement,
+            "source_display_order": [
+                "source_country", "source_company", "execution_method",
+                "internet_requirement", "model_name"
+            ],
+            # 5.2+ rich fields (non-breaking; frontend + backend read if present)
+            "hf_repo_id": self.hf_repo_id,
+            "quantization": self.quantization,
+            "download_strategy": self.download_strategy,
+            "load_strategy": self.load_strategy,
+            "hardware": {
+                "min_ram_gb": self.hardware.min_ram_gb,
+                "recommended_ram_gb": self.hardware.recommended_ram_gb,
+                "apple_silicon_pref": self.hardware.apple_silicon_pref,
+                "cuda_pref": self.hardware.cuda_pref,
+                "notes": self.hardware.notes,
+            },
+            "license": self.license,
+            "safety_notes": self.safety_notes,
+            "verification": {
+                "hf_exists": self.verification.hf_exists,
+                "hf_last_checked": self.verification.hf_last_checked,
+                "has_config": self.verification.has_config,
+                "has_tokenizer": self.verification.has_tokenizer,
+                "has_weights_hint": self.verification.has_weights_hint,
+                "pipeline_tag": self.verification.pipeline_tag,
+                "verified": bool(
+                    self.verification.hf_exists
+                    and self.verification.has_config
+                    and self.verification.has_tokenizer
+                ),
+                "notes": self.verification.notes,
+            },
+            "recommended_default": self.recommended_default,
+        }
+        return base
+
+
+# ── Curated 5.2.0 registry (bold user-focused: transparent, multimodal-first, verified where practical) ──
+# Current Gemma-4 / Qwen3-VL / Llama-4 kept + modern additions (Gemma3, Qwen2.5-VL, Llama-3.2-Vision, Pixtral).
+# Presence verified via HF API (lightweight model_info) on 2026-06-14.
+# Full weight download is user-consent only; entries without config/tokenizer
+# hints are shown as available-but-not-load-verified.
+
+_REGISTRY: List[ModelCapability] = [
+    # Gemma 4 family (mlx-community 4-bit, Apple-first, excellent local VLM)
+    ModelCapability(
+        id="mlx-community/gemma-4-e2b-4bit",
+        hf_repo_id="mlx-community/gemma-4-e2b-4bit",
+        name="Gemma 4 E2B Base",
+        family="Gemma 4",
+        tag="local-vlm",
+        size="3.6GB",
+        quantization="4bit",
+        provider_hints=["local_mlx"],
+        download_strategy="hf_hub",
+        load_strategy="mlx_vlm",
+        hardware=HardwareProfile(min_ram_gb=6.0, recommended_ram_gb=8.0, apple_silicon_pref=True, notes="Tiny but capable vision; great first local VLM."),
+        source_country="미국", source_company="Google",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="any-to-any", license="apache-2.0", verified_by="hf-api-light"),
+        recommended_default=True,
+        display_priority=10,
+    ),
+    ModelCapability(
+        id="mlx-community/gemma-4-e2b-it-4bit",
+        hf_repo_id="mlx-community/gemma-4-e2b-it-4bit",
+        name="Gemma 4 E2B Instruct",
+        family="Gemma 4",
+        tag="local-vlm",
+        size="3.6GB",
+        quantization="4bit",
+        provider_hints=["local_mlx"],
+        download_strategy="hf_hub",
+        load_strategy="mlx_vlm",
+        hardware=HardwareProfile(min_ram_gb=6.0, recommended_ram_gb=8.0, apple_silicon_pref=True, notes="Instruct-tuned; preferred over base for chat."),
+        source_country="미국", source_company="Google",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="any-to-any", license="apache-2.0", verified_by="hf-api-light"),
+        recommended_default=True,
+        display_priority=11,
+    ),
+    ModelCapability(
+        id="mlx-community/gemma-4-e4b-4bit",
+        hf_repo_id="mlx-community/gemma-4-e4b-4bit",
+        name="Gemma 4 E4B Base",
+        family="Gemma 4",
+        tag="local-vlm",
+        size="5.2GB",
+        quantization="4bit",
+        provider_hints=["local_mlx"],
+        download_strategy="hf_hub",
+        load_strategy="mlx_vlm",
+        hardware=HardwareProfile(min_ram_gb=8.0, recommended_ram_gb=10.0, apple_silicon_pref=True),
+        source_country="미국", source_company="Google",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="any-to-any", license="apache-2.0", verified_by="hf-api-light"),
+    ),
+    ModelCapability(
+        id="mlx-community/gemma-4-e4b-it-4bit",
+        hf_repo_id="mlx-community/gemma-4-e4b-it-4bit",
+        name="Gemma 4 E4B Instruct",
+        family="Gemma 4",
+        tag="local-vlm",
+        size="5.2GB",
+        quantization="4bit",
+        provider_hints=["local_mlx"],
+        download_strategy="hf_hub",
+        load_strategy="mlx_vlm",
+        hardware=HardwareProfile(min_ram_gb=8.0, recommended_ram_gb=10.0, apple_silicon_pref=True),
+        source_country="미국", source_company="Google",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="any-to-any", license="apache-2.0", verified_by="hf-api-light"),
+    ),
+    ModelCapability(
+        id="mlx-community/gemma-4-12b-it-4bit",
+        hf_repo_id="mlx-community/gemma-4-12b-it-4bit",
+        name="Gemma 4 12B Instruct",
+        family="Gemma 4",
+        tag="local-vlm",
+        size="7.6GB",
+        quantization="4bit",
+        provider_hints=["local_mlx", "ollama", "vllm", "lmstudio", "llamacpp"],
+        download_strategy="hf_hub",
+        load_strategy="mlx_vlm",
+        hardware=HardwareProfile(min_ram_gb=12.0, recommended_ram_gb=16.0, apple_silicon_pref=True, notes="Sweet spot for local multimodal on M-series 16GB+ or 24GB+."),
+        source_country="미국", source_company="Google",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="image-text-to-text", license="apache-2.0", verified_by="hf-api-light"),
+        recommended_default=True,
+        display_priority=20,
+    ),
+    ModelCapability(
+        id="mlx-community/gemma-4-26b-a4b-it-4bit",
+        hf_repo_id="mlx-community/gemma-4-26b-a4b-it-4bit",
+        name="Gemma 4 26B A4B Instruct",
+        family="Gemma 4",
+        tag="local-vlm",
+        size="15.6GB",
+        quantization="4bit",
+        provider_hints=["local_mlx"],
+        download_strategy="hf_hub",
+        load_strategy="mlx_vlm",
+        hardware=HardwareProfile(min_ram_gb=20.0, recommended_ram_gb=28.0, apple_silicon_pref=True, notes="Large MoE-style; local load practical only on high-RAM Apple Silicon (32GB+). Long download expected."),
+        source_country="미국", source_company="Google",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="image-text-to-text", license="apache-2.0", verified_by="hf-api-light"),
+        display_priority=50,
+    ),
+    ModelCapability(
+        id="mlx-community/gemma-4-31b-it-4bit",
+        hf_repo_id="mlx-community/gemma-4-31b-it-4bit",
+        name="Gemma 4 31B Instruct",
+        family="Gemma 4",
+        tag="local-vlm",
+        size="18.4GB",
+        quantization="4bit",
+        provider_hints=["local_mlx", "ollama", "vllm"],
+        download_strategy="hf_hub",
+        load_strategy="mlx_vlm",
+        hardware=HardwareProfile(min_ram_gb=24.0, recommended_ram_gb=32.0, apple_silicon_pref=True, notes="Very large; high-end local only. Consider cloud fallback for lower RAM."),
+        source_country="미국", source_company="Google",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="image-text-to-text", license="apache-2.0", verified_by="hf-api-light"),
+    ),
+
+    # Qwen3-VL (strong real-world multimodal, good small sizes)
+    ModelCapability(
+        id="mlx-community/Qwen3-VL-4B-Instruct-4bit",
+        hf_repo_id="mlx-community/Qwen3-VL-4B-Instruct-4bit",
+        name="Qwen3-VL 4B",
+        family="Qwen3-VL",
+        tag="local-vlm",
+        size="2.7GB",
+        quantization="4bit",
+        provider_hints=["local_mlx", "ollama", "vllm", "lmstudio", "llamacpp"],
+        download_strategy="hf_hub",
+        load_strategy="mlx_vlm",
+        hardware=HardwareProfile(min_ram_gb=5.0, recommended_ram_gb=8.0, apple_silicon_pref=True, notes="Extremely compact strong VLM. Best default for low-RAM Macs."),
+        source_country="중국", source_company="Alibaba",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="image-text-to-text", license="apache-2.0", verified_by="hf-api-light"),
+        recommended_default=True,
+        display_priority=5,
+    ),
+    ModelCapability(
+        id="mlx-community/Qwen3-VL-8B-Instruct-4bit",
+        hf_repo_id="mlx-community/Qwen3-VL-8B-Instruct-4bit",
+        name="Qwen3-VL 8B",
+        family="Qwen3-VL",
+        tag="local-vlm",
+        size="4.8GB",
+        quantization="4bit",
+        provider_hints=["local_mlx", "ollama", "vllm", "lmstudio", "llamacpp"],
+        download_strategy="hf_hub",
+        load_strategy="mlx_vlm",
+        hardware=HardwareProfile(min_ram_gb=8.0, recommended_ram_gb=12.0, apple_silicon_pref=True),
+        source_country="중국", source_company="Alibaba",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="image-text-to-text", license="apache-2.0", verified_by="hf-api-light"),
+        recommended_default=True,
+        display_priority=15,
+    ),
+    ModelCapability(
+        id="mlx-community/Qwen3-VL-30B-A3B-Instruct-4bit",
+        hf_repo_id="mlx-community/Qwen3-VL-30B-A3B-Instruct-4bit",
+        name="Qwen3-VL 30B A3B",
+        family="Qwen3-VL",
+        tag="local-vlm",
+        size="18GB",
+        quantization="4bit",
+        provider_hints=["local_mlx", "ollama", "vllm"],
+        download_strategy="hf_hub",
+        load_strategy="mlx_vlm",
+        hardware=HardwareProfile(min_ram_gb=24.0, recommended_ram_gb=32.0, apple_silicon_pref=True, notes="Large MoE VLM; practical local only on 32GB+ Apple Silicon or strong CUDA. Download is multi-GB."),
+        source_country="중국", source_company="Alibaba",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="image-text-to-text", license="apache-2.0", verified_by="hf-api-light"),
+    ),
+
+    # Llama 4
+    ModelCapability(
+        id="mlx-community/Llama-4-Scout-17B-16E-Instruct-4bit",
+        hf_repo_id="mlx-community/Llama-4-Scout-17B-16E-Instruct-4bit",
+        name="Llama 4 Scout 17B 16E",
+        family="Llama 4",
+        tag="local-vlm",
+        size="11.8GB",
+        quantization="4bit",
+        provider_hints=["local_mlx", "ollama", "vllm", "lmstudio", "llamacpp"],
+        download_strategy="hf_hub",
+        load_strategy="mlx_vlm",
+        hardware=HardwareProfile(min_ram_gb=16.0, recommended_ram_gb=20.0, apple_silicon_pref=True),
+        source_country="미국", source_company="Meta",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="image-text-to-text", license="llama3.1-ish / meta-llama", verified_by="hf-api-light"),
+        recommended_default=True,
+        display_priority=25,
+    ),
+
+    # ── Modern additions for 5.2.0 (verified on HF, user choice expansion) ──
+    # Gemma 3 (excellent real multimodal balance, smaller than 4 where present)
+    ModelCapability(
+        id="google/gemma-3-4b-it",
+        hf_repo_id="google/gemma-3-4b-it",
+        name="Gemma 3 4B Instruct (HF)",
+        family="Gemma 3",
+        tag="local-vlm",
+        size="~5GB+",
+        quantization="bf16 / 4bit variants",
+        provider_hints=["local_mlx", "vllm", "ollama"],
+        download_strategy="hf_hub",
+        load_strategy="mlx_vlm",
+        hardware=HardwareProfile(min_ram_gb=8.0, recommended_ram_gb=12.0, apple_silicon_pref=True, notes="Use mlx-community quantized ports when available for best local perf."),
+        source_country="미국", source_company="Google",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="image-text-to-text", license="gemma-terms", verified_by="hf-api-light"),
+        display_priority=30,
+    ),
+    ModelCapability(
+        id="google/gemma-3-12b-it",
+        hf_repo_id="google/gemma-3-12b-it",
+        name="Gemma 3 12B Instruct (HF)",
+        family="Gemma 3",
+        tag="local-vlm",
+        size="~12GB+",
+        quantization="bf16 / GGUF-4bit",
+        provider_hints=["ollama", "vllm", "lmstudio", "llamacpp"],
+        download_strategy="hf_hub",
+        load_strategy="ollama",
+        hardware=HardwareProfile(min_ram_gb=16.0, recommended_ram_gb=20.0, notes="Prefer quantized GGUF for llama.cpp / ollama on non-Apple or lower RAM."),
+        source_country="미국", source_company="Google",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="image-text-to-text", license="gemma-terms", verified_by="hf-api-light"),
+    ),
+
+    # Qwen2.5-VL (battle-tested, widely supported)
+    ModelCapability(
+        id="Qwen/Qwen2.5-VL-7B-Instruct",
+        hf_repo_id="Qwen/Qwen2.5-VL-7B-Instruct",
+        name="Qwen2.5-VL 7B Instruct",
+        family="Qwen2.5-VL",
+        tag="local-vlm",
+        size="~8-15GB (quant dependent)",
+        quantization="AWQ / GGUF / 4bit ports",
+        provider_hints=["vllm", "ollama", "lmstudio", "llamacpp"],
+        download_strategy="hf_hub",
+        load_strategy="vllm",
+        hardware=HardwareProfile(min_ram_gb=12.0, recommended_ram_gb=16.0, cuda_pref=True, notes="Strong general VLM. mlx-community or GGUF ports recommended for local Apple."),
+        source_country="중국", source_company="Alibaba",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="image-text-to-text", license="apache-2.0", verified_by="hf-api-light"),
+        display_priority=35,
+    ),
+
+    # Llama 3.2 Vision (widely available, good ecosystem)
+    ModelCapability(
+        id="meta-llama/Llama-3.2-11B-Vision-Instruct",
+        hf_repo_id="meta-llama/Llama-3.2-11B-Vision-Instruct",
+        name="Llama 3.2 11B Vision Instruct",
+        family="Llama 3.2 Vision",
+        tag="local-vlm",
+        size="~11-22GB (quant)",
+        quantization="Q4_K_M GGUF widely available",
+        provider_hints=["ollama", "llamacpp", "lmstudio", "vllm"],
+        download_strategy="hf_hub",
+        load_strategy="ollama",
+        hardware=HardwareProfile(min_ram_gb=14.0, recommended_ram_gb=18.0, notes="Excellent GGUF support. Ollama / llama.cpp default path for most users."),
+        source_country="미국", source_company="Meta",
+        verification=VerificationStatus(hf_exists=True, has_config=True, has_tokenizer=True, has_weights_hint=True, pipeline_tag="image-text-to-text", license="llama3.2", verified_by="hf-api-light"),
+        display_priority=40,
+    ),
+
+    # Pixtral (Mistral multimodal, strong)
+    ModelCapability(
+        id="mistralai/Pixtral-12B-2409",
+        hf_repo_id="mistralai/Pixtral-12B-2409",
+        name="Pixtral 12B (Mistral)",
+        family="Pixtral",
+        tag="local-vlm",
+        size="~12-24GB",
+        quantization="GGUF / AWQ ports",
+        provider_hints=["vllm", "ollama", "lmstudio"],
+        download_strategy="hf_hub",
+        load_strategy="vllm",
+        hardware=HardwareProfile(min_ram_gb=16.0, recommended_ram_gb=20.0, cuda_pref=True, notes="High quality vision-language. Best on CUDA / vLLM; GGUF for CPU/Apple via community ports."),
+        source_country="프랑스", source_company="Mistral AI",
+        verification=VerificationStatus(
+            hf_exists=True,
+            has_config=False,
+            has_tokenizer=False,
+            has_weights_hint=True,
+            pipeline_tag=None,
+            license="mistral-research",
+            notes="HF repo and weights are present, but config/tokenizer files were not visible in the lightweight HF tree check; treat as available but not local-load verified.",
+            verified_by="hf-api-light",
+        ),
+        display_priority=45,
+    ),
+]
+
+
+def get_all_capabilities() -> List[ModelCapability]:
+    return list(_REGISTRY)
+
+
+def get_capability(model_id: str) -> Optional[ModelCapability]:
+    for m in _REGISTRY:
+        if m.id == model_id or m.hf_repo_id == model_id:
+            return m
+    return None
+
+
+def build_engine_model_catalog() -> Dict[str, List[Dict[str, Any]]]:
+    """Return legacy ENGINE_MODEL_CATALOG shape, enriched with 5.2 fields."""
+    from collections import defaultdict
+    by_engine: Dict[str, List[Dict[str, Any]]] = defaultdict(list)
+
+    engine_map = {
+        "local_mlx": ["local_mlx"],
+        "ollama": ["ollama"],
+        "vllm": ["vllm"],
+        "lmstudio": ["lmstudio"],
+        "llamacpp": ["llamacpp"],
+    }
+
+    for cap in _REGISTRY:
+        for eng_key, hints in engine_map.items():
+            if any(h in cap.provider_hints for h in hints) or eng_key in cap.provider_hints:
+                legacy = cap.to_legacy_dict()
+                # Adapt id for non-mlx engines (match historical patterns)
+                if eng_key == "ollama" and not legacy["id"].startswith("ollama:"):
+                    # historical used prefixed or hf.co for some
+                    if "gguf" in cap.tag.lower() or "gguf" in (cap.quantization or "").lower():
+                        legacy["id"] = f"ollama:hf.co/ggml-org/{cap.family.lower().replace(' ', '')}-12B-it-GGUF:Q4_K_M"  # fallback, overridden by aliases
+                    else:
+                        legacy["id"] = f"ollama:{cap.hf_repo_id.split('/')[-1].lower()}"
+                elif eng_key == "vllm" and not legacy["id"].startswith("vllm:"):
+                    legacy["id"] = f"vllm:{cap.hf_repo_id}"
+                elif eng_key == "lmstudio" and not legacy["id"].startswith("lmstudio:"):
+                    legacy["id"] = f"lmstudio:{cap.hf_repo_id}"
+                elif eng_key == "llamacpp" and not legacy["id"].startswith("llamacpp:"):
+                    legacy["id"] = f"llamacpp:{cap.hf_repo_id}-GGUF"
+                by_engine[eng_key].append(legacy)
+
+    # Ensure at least the primary local_mlx ones are present (exact historical)
+    # If projection missed any, inject the original local_mlx entries enriched
+    if not by_engine.get("local_mlx"):
+        for cap in _REGISTRY:
+            if "local_mlx" in cap.provider_hints:
+                by_engine["local_mlx"].append(cap.to_legacy_dict())
+
+    return {k: v for k, v in by_engine.items()}
+
+
+def get_verified_models() -> List[Dict[str, Any]]:
+    """Return only load-verified HF entries with rich fields (for API/UI)."""
+    return [
+        c.to_legacy_dict() for c in _REGISTRY
+        if c.verification.hf_exists and c.verification.has_config and c.verification.has_tokenizer
+    ]
+
+
+# Back-compat: expose a simple list mirroring the old top-level for mlx
+LOCAL_MLX_MODELS = [c.to_legacy_dict() for c in _REGISTRY if "local_mlx" in c.provider_hints]