ModelIntegrity.dev gives engineers and compliance teams a single source of truth for evaluating open-weight LLMs — from hardware costs to regulatory clearance.
23+
Open-Weight LLMs
13+
Local SLMs
6
Compliance Frameworks
10
Benchmark Suites
From initial model selection to production compliance auditing — everything an ML team needs in one platform.
22+ open-weight LLMs with full architectural specs — attention heads, layer dimensions, quantization support, and training data provenance.
Automated scoring against HIPAA, NIST AI RMF, EU AI Act, SOC 2, GDPR, and ISO 42001. Know exactly which models are cleared for production.
Cross-model performance on MMLU, HumanEval, MATH, GSM8K, TruthfulQA, GPQA Diamond, and more — all in one visual dashboard.
Interactive VRAM calculator for any quantization level and context window. Linked directly to real AWS, GCP, and Azure instance pricing.
Specialized agents for hardware configuration advisory and technical spec validation — ask in plain English, get infrastructure-grade answers.
Programmatic access to model specs, compliance scores, and benchmark data. Integrate ModelIntegrity into your own MLOps toolchain.
12+ small language models that run on a laptop, Raspberry Pi, or even a browser — no GPU required. From Phi-4 to Gemma 4, with Ollama quick-start commands.
Three steps from model discovery to production-ready deployment.
Search 22+ open-weight models by architecture, parameter size, or developer. Compare any two side by side.
Run automated scoring against your required regulatory frameworks. Get a pass/fail status per model, per framework.
Calculate exact VRAM requirements, get hardware recommendations, and see real cloud costs before you commit.
No sign-up required to explore the catalog. Start browsing models, running compliance checks, and planning hardware today.
Get Started — It's Free