Context
262K
Input
$0.05/M
Output
$0.20/M
Aliases:nemotron-lightningnemotron-lightning-3.5nemotron-lightning-3.5-30b-a3bnvidia-nemotron-lightning-3.5-30b-a3b
NVIDIA's hybrid Mamba-Transformer MoE model (31B total, 3B active) with a multi-token prediction speculative decoding head for low-latency serving. Tuned for high-throughput reasoning and agentic workloads.
crusoe/nemotron-lightning-3.5-30bQuantUnspecified
- Input
- $0.05/M
- Output
- $0.20/M
- Context
- 262K
- Max out
- 66K
- Cache read
- $0.03/M
- Cache write
- $0.05/M
- Thinking out
- $0.20/M