Nvidia Releases Nemotron 3 Embed 8B Model Topping RTEB Benchmark

aiai-modelingai-model-releasesai-open-modelsai-research-evals 3 posts · 3 accounts

Nvidia released the Nemotron 3 Embed 8B parameters multilingual retrieval model, which took the top overall position on the RTEB benchmark. The launch includes a companion 1B parameters embedding model optimized for smaller-scale tasks.

RTEB measures retrieval accuracy across real-world tasks, providing AI agents with relevant context to improve response accuracy. Both variants are licensed under Open Model Wallet 1.1 and are available for commercial use.

From the sources (3 posts)

@nvidiaai

Today we released Nemotron 3 Embed 8B and it reached #1 overall on RTEB 🏆 RTEB benchmarks retrieval accuracy across real-world tasks. Better retrieval gives agents more relevant context, helping improve response accuracy.

@tomaarsen

I'm very excited to share that @nvidia just released Nemotron-3-Embed: two multilingual embedding models for retrieval The 8B takes the top spot on RTEB. - Nemotron-3-Embed-1B-BF16 (2048-dim) - Nemotron-3-Embed-8B-BF16 (4096-dim) Both Op

@michaelgannotti

NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB, Advancing Agentic Retrieval @NVIDIAAI

FDA Recalls More Than 735,000 Pillsbury Bread Rolls Over Glass Contamination Nvidia Releases ARDY, an Open-Source AI Animation Model Generating Motion on 24GB GPUs