InfoCapability
NVIDIA AI-Q achieves top rank on DeepResearch Bench using Llama Nemotron
AI Impact Summary
NVIDIA’s AI-Q utilizes Llama-3.3-Nemotron-Super-49B-v1.5 and Llama 3.3-70B Instruct models to achieve top performance on the DeepResearch Bench, specifically within the LLM with Search category. This demonstrates the potential of open-source models for complex agentic workflows, including multi-step reasoning and tool use, and represents a significant advancement for the open-source AI stack.
NVIDIA AI-Q achieves top rank on DeepResearch Bench using Llama Nemotron
Affected Systems
Llama 3.3-70B InstructLlama-3.3-Nemotron-Super-49B-v1.5
- Date
- Date not specified
- Change type
- capability
- Severity
- info