About Non finito
Non finito is a platform designed for evaluating multimodal models, with a focus on making it easy to run and share evaluations. It aims to provide tools specifically tailored for multimodal models, which are often overlooked by tools primarily focused on language models (LLMs). The platform emphasizes easy comparison of models and public sharing of evaluations.
How to Use
Users can sign up to create evaluations, compare different models, view public evaluations, and access examples to understand the platform's capabilities. The platform allows input and output comparisons for various tasks.
Non finito's
Key Features
- Multimodal model evaluation
- Model comparison
- Public sharing of evaluations
- Example evaluations for various tasks
Use Cases
- Evaluating entity tracking in language models
- Evaluating logical reasoning capabilities
- Real-world question answering
- Visual deductive reasoning
- Evaluating visual models by counting cards
- Graph comprehension and policy recommendation
Key Features
Multimodal model evaluation
Model comparison
Public sharing of evaluations
Example evaluations for various tasks
Best For
Evaluating entity tracking in language modelsEvaluating logical reasoning capabilitiesReal-world question answeringVisual deductive reasoningEvaluating visual models by counting cardsGraph comprehension and policy recommendation