Reference

LLM models

Explore model specifications, benchmarks and costs.

Catalog snapshot455 models · 53 labs

Reference

LLM models

Explore model specifications, benchmarks and costs.

455 models

Nvidia

Llama Nemotron Embed VL 1B v2

llama-nemotron-embed-vl-1b-v2
Context
32768
Max output
2048
Released
2026-02-10
Weights
Open weights

API pricing

No exact-version pricing recorded yet.

Sources

Back to results