Our growing collection of task specific LLM as a Judge adapters for performing evaluations
AI & ML interests
As part of our mission to foster grounded generative AI applications, we have trained and released SLMs tailored to model based evaluation. Our open source models aim to provide efficient, high-performing alternatives to black box LLMs currently used for evaluating GAI apps, ensuring outputs remain grounded in factual information and ethical principles.
Recent Activity
Base model(s) merged with the specific evaluation task adapter. Each model performs excellently for its purpose and remains useful for general tasks.
-
grounded-ai/phi3-toxicity-judge-merge
Text Generation • 4B • Updated • 18 -
grounded-ai/phi3-rag-relevance-judge-merge
Text Generation • 4B • Updated • 19 • 1 -
grounded-ai/phi3-hallucination-judge-merge
Text Generation • 4B • Updated • 16 • 1 -
grounded-ai/phi3.5-hallucination-judge-merge
Text Generation • 4B • Updated • 19 • 2
Our growing collection of task specific LLM as a Judge adapters for performing evaluations
Base model(s) merged with the specific evaluation task adapter. Each model performs excellently for its purpose and remains useful for general tasks.
-
grounded-ai/phi3-toxicity-judge-merge
Text Generation • 4B • Updated • 18 -
grounded-ai/phi3-rag-relevance-judge-merge
Text Generation • 4B • Updated • 19 • 1 -
grounded-ai/phi3-hallucination-judge-merge
Text Generation • 4B • Updated • 16 • 1 -
grounded-ai/phi3.5-hallucination-judge-merge
Text Generation • 4B • Updated • 19 • 2