Nemotron 3.5 Bolt 74B A9B
Parameters
74B
Context Length
-
Modality
Text
Architecture
-
KV Cache Rate
128 KB/tok
License
other
Release Date
7 Oct 2026
Knowledge Cutoff
-
API Pricing (per 1M)
Self-hosted only
Hardware Sizing
Precision
Context
Workload Memory: 141.34 GB
FP16 · 8K context
Base Weights: 137.84 GB
KV Cache: 1.00 GB
Runtime & Graphs: 2.50 GB
Minimum Required Hardware
2x NVIDIA H100 80GB
141.3 / 144.0 GB usable · TP 2
8x RTX 4090 24GB
141.3 / 172.8 GB usable · TP 8
Apple M2 Ultra 192GB
141.3 / 144.0 GB usable
Determine requirements for your specific workload:
Open in VRAM Calculator
Architecture Diagram
Evaluation Benchmarks
Rank
#102
| Benchmark | Score | Rank |
|---|---|---|
Intelligence Index | 0.21 | 148 |
Rankings
Overall Rank
#102
Coding Rank
-
About Nemotron 3.5 Bolt 74B A9B
Nemotron 3.5 Bolt 74B A9B is a high-speed multimodal MoE model engineered by NVIDIA for fast reasoning and visual comprehension within a 24GB memory footprint. It utilizes dynamic routing to activate 9B parameters per token for low-latency serving.
Technical Specifications
Resources
About Nemotron 3.5
NVIDIA's Nemotron 3.5 family of open models delivers leading efficiency and accuracy for agentic AI applications. Built on an advanced hybrid Mamba-2 SSM and sparse Mixture-of-Experts architecture with native Multi-Token Prediction (MTP) and speculative decoding harnesses, Nemotron 3.5 models are purpose-built for the high-volume execution layer of always-on agents.