D
Nemotron 3.5 Rerank 8B
Nemotron 3.5 Rerank 8B, a reranking model from DeepInfra.
model: "nemotron-3.5-rerank-8b"
NNvidia·served viaDeepInfra·Released October 2026
Pricing
- Per 1M tokens
- $0.050 / 1M
Specifications
- Context window
- 32,768 tokens
- Max documents
- 1,000
- Max document tokens
- 32,768
- Released
- October 2026
- Model string
- nvidia/Nemotron-3.5-Rerank-8B-BF16
Use this model
// Reranking runs inside vector search, reordering results // by relevance. Selected in the corpus configuration. // Model id: 0