Reference — Documentation & evaluations

System Cards & Model Documentation

Where to find the primary documentation for the models in the registry: system cards, model cards, technical reports and independent evaluations. These are the documents that answer governance questions — what a model was trained on, how it was evaluated, what it refuses, and what the vendor claims versus what independents measured.

Reading the document types

System card
Describes a deployed system — model plus safeguards, evaluations, red-teaming results and known limitations at release. The frontier labs publish one per major release; it is the closest thing to a safety datasheet and the first document to pull for a risk assessment.
Model card
Describes the model artifact itself — intended use, training summary, benchmarks, license. Standard for open-weight releases, usually on the Hugging Face repository alongside the weights.
Technical report
Architecture, training method and evaluation detail. Depth varies enormously: some vendors publish data recipes, others a benchmark table.

For EU deployments these documents map directly onto obligations: the AI Act's general-purpose AI rules (applicable since August 2025) require providers to maintain technical documentation and training-data summaries, and downstream deployers inherit the need to understand them. See the Commission's AI Act framework.

Frontier labs — system cards per release

Proprietary vendors publish safety and system documentation per model release, at varying depth.

Anthropic 🇺🇸 USA

System cards for each Claude release, plus a transparency hub covering safety framework, evaluations and government commitments. Among the most detailed system cards published.

OpenAI 🇺🇸 USA

System cards per release (GPT-5 onward under the unified safety framework) plus preparedness framework reports and the safety hub.

Google DeepMind 🇺🇸 USA

Model pages with technical reports and model/system cards for the Gemini family, plus the Frontier Safety Framework documentation.

Microsoft AI 🇺🇸 USA

Model pages for the MAI family; the MAI-Thinking-1 technical report also covers the lineage behind MAI-Cyber-1-Flash.

Meta 🇺🇸 USA

Developer documentation for the Muse family and legacy Llama model cards. Note the documentation split after the shift from open-weight Llama to proprietary Muse.

xAI 🇺🇸 USA

Model announcements and documentation for the Grok family; safety documentation is comparatively thin — worth noting in assessments.

Thinking Machines Lab 🇺🇸 USA

The Inkling announcement doubles as a technical report — architecture, training data scale, safety benchmarks (FORTRESS) and calibration methodology.

Open-weight vendors — model cards with the weights

For open models the primary documentation lives on the Hugging Face organisation pages, next to the checkpoints.

DeepSeek 🇨🇳 China

Model cards and technical reports for the V4 family and predecessors, published with MIT-licensed weights.

Alibaba Qwen 🇨🇳 China

Cards for the Qwen 3.x open-weight line, including the 3.6-generation MoE models.

Z.ai (Zhipu) 🇨🇳 China

GLM-5.x model cards and agentic evaluation detail, MIT-licensed.

Moonshot AI 🇨🇳 China

Kimi model cards; check here for the K3 open-weight checkpoint status.

Xiaomi MiMo 🇨🇳 China

MiMo-V2.5-Pro card with training methodology and agentic benchmark detail.

StepFun 🇨🇳 China

Step 3.7 Flash cards including quantised deployment recipes for local use.

NVIDIA 🇺🇸 USA

Nemotron 3 model cards — unusual for including the training datasets themselves, not just their description.

Mistral AI 🇫🇷 France

Model cards and API documentation for the open and commercial lines — the EU-jurisdiction reference vendor.

IBM Granite 🇺🇸 USA

Granite 4 cards with the governance extras: signed weights and ISO/IEC 42001-accredited process documentation.

Cisco Foundation AI 🇺🇸 USA

Cards for the security models — Foundation-Sec-8B and the Antares family (gated access).

Model-card standard 🌐

Hugging Face's documentation of the model-card format itself — useful when writing internal cards for fine-tuned models.

Independent evaluation & tracking

Vendor documentation states claims; these sources test or track them independently — always pair the two.

UK AI Security Institute 🇬🇧 UK

Government pre-deployment evaluations of frontier models, including the cyber-range exercises referenced for Mythos-class models.

Artificial Analysis 🌐

Independent intelligence, speed and price benchmarking across hosted models — the fastest way to sanity-check a vendor benchmark table.

Stanford HELM 🇺🇸 USA

Academic holistic evaluation framework with transparent, reproducible scenario coverage.

LMArena 🌐

Blind human-preference rankings — noisy but hard to game in the ways static benchmarks are.

Epoch AI 🌐

Data and analysis on compute, parameters and training trends — the source for scale context behind the registry's log axis.

models.dev & OpenRouter 🌐

Machine-readable model specification and pricing databases — practical sync sources for keeping the registry's pricing fields current.