Foundation Model Transparency Index
by Stanford CRFMPublished 2025-12-01Source ↗Overall:14
- Agent Protocols
- 0
- AI bug bounty
- 0
- Amount of usage
- 0
- AUP enforcement frequency
- 0
- AUP enforcement process
- 100
- Basic model properties
- 0
- Benchmarked inference
- 0
- Benefits Assessment
- 0
- Capabilities evaluation
- 0
- Capabilities taxonomy
- 0
- Carbon emissions for final training run
- 0
- Change log
- 100
- Classification of usage data
- 0
- Code access
- 0
- Compute hardware for final training run
- 0
- Compute provider
- 0
- Compute usage for final training run
- 0
- Compute usage including R&D
- 0
- Consumer/enterprise usage
- 0
- Crawling
- 0
- Data acquisition methods
- 0
- Data domain composition
- 0
- Data laborer practices
- 0
- Data language composition
- 0
- Data processing methods
- 0
- Data processing purpose
- 0
- Data processing techniques
- 0
- Data replicability
- 0
- Data retention and deletion policy
- 0
- Data size
- 0
- Deeper model properties
- 0
- Detection of machine-generated content
- 0
- Development duration for final training run
- 0
- Distribution channels with usage data
- 100
- Documentation for responsible use
- 100
- Downstream
- 25
- Acceptable use policy
- 60
- Accountability
- 0
- Downstream mitigations
- 40
- Impact
- 28.6
- Model Behavior Policy
- 25
- Post-deployment monitoring
- 0
- Usage data
- 20
- Energy usage for final training run
- 0
- Enterprise mitigations
- 0
- Enterprise users
- 0
- External data access
- 0
- External developer mitigations
- 100
- External products and services
- 100
- External reproducibility of capabilities evaluation
- 0
- External reproducibility of mitigations evaluation
- 0
- External reproducibility of risks evaluation
- 0
- External risk evaluation
- 0
- Feedback mechanisms
- 0
- Foundation model roadmap
- 100
- Geographic statistics
- 0
- Government commitments
- 0
- Government use
- 0
- Instructions for data generation
- 0
- Intermediate tokens
- 0
- Internal compute allocation
- 0
- Internal product and service mitigations
- 0
- Internal products and services
- 100
- Licensed data compensation
- 0
- Licensed data sources
- 0
- Misuse incident reporting protocol
- 0
- Mitigations efficacy
- 0
- Mitigations taxonomy
- 0
- Mitigations taxonomy mapped to risk taxonomy
- 0
- Model
- 16.7
- Capabilities
- 0
- Model cost
- 0
- Model dependencies
- 0
- Model access
- 0
- Model information
- 0
- Model Mitigations
- 0
- Model objectives
- 0
- Release
- 62.5
- Model response characteristics
- 100
- Risks
- 0
- Model stages
- 0
- Model theft prevention measures
- 0
- New human-generated data sources
- 0
- Notice of usage data used in training
- 0
- Open weights
- 0
- Organization chart
- 0
- Oversight mechanism
- 0
- Permitted and prohibited users
- 100
- Permitted, restricted, and prohibited model behaviors
- 0
- Permitted, restricted, and prohibited uses
- 100
- Post-deployment coordination with government
- 0
- Pre-deployment risk evaluation
- 0
- Public datasets
- 0
- Quantization
- 100
- Regional policy variations
- 0
- Release stages
- 0
- Researcher credits
- 0
- Responsible disclosure policy
- 0
- Risks evaluation
- 0
- Risks taxonomy
- 0
- Risk thresholds
- 0
- Safe harbor
- 0
- Security incident reporting protocol
- 0
- Specialized access
- 0
- Synthetic data purpose
- 0
- Synthetic data sources
- 0
- System prompt
- 0
- Terms of use
- 100
- Top distribution channels
- 100
- Train-test overlap
- 0
- Upstream
- 0
- Compute
- 0
- Data Acquisition
- 0
- Data Processing
- 0
- Data Properties
- 0
- Methods
- 0
- Other resources
- 0
- Usage data used in training
- 0
- Users of internal products and services
- 0
- Versioning protocol
- 0
- Water usage for final training run
- 0
- Whistleblower protection
- 0