How much VRAM does Nemotron 3.5 Bolt 74B A9B need?
Nemotron 3.5 Bolt 74B A9B needs 42.3 GB for its weights at 4-bit, 43.3 GB with an 8K context and 43.5 GB with 32K. At 8-bit, 75.2 GB.
- Parameters
- 74.9B
- Active per token
- 9B
- Longest context
- 256K
- Released
- Oct 2026
- License
- OpenMDW 1.1
- Strongest at
- Chat, Coding, Agents, Docs
Memory at each precision
| Precision | Weights | With 8K | With 32K | With 128K |
|---|---|---|---|---|
| 16-bit | 139 GB | 141 GB | 141 GB | 141 GB |
| 8-bit | 74.1 GB | 75.2 GB | 75.3 GB | 76.1 GB |
| 6-bit | 57.2 GB | 58.2 GB | 58.4 GB | 59.2 GB |
| 5-bit | 49.6 GB | 50.7 GB | 50.8 GB | 51.6 GB |
| 4-bit | 42.3 GB | 43.3 GB | 43.5 GB | 44.3 GB |
| 3-bit | 34.1 GB | 35.1 GB | 35.3 GB | 36.1 GB |
| 2-bit | 26.2 GB | 27.2 GB | 27.4 GB | 28.1 GB |
| 1.6-bit | 14.4 GB | 15.4 GB | 15.6 GB | 16.4 GB |
One conversation, 16-bit context cache, runtime overhead included.
What runs it
| Machine | Best fit | Context | Speed | First reply |
|---|---|---|---|---|
| RTX 3060 12 GB | Too big | |||
| RTX 4060 Ti 16 GB | Too big | |||
| RTX 3090 | Too big | |||
| RTX 4090 | Too big | |||
| RTX 5090 | Too big | |||
| 2× RTX 3090 | 4-bit | 216K | 72 tok/s | 1.0 s |
| RTX PRO 6000 Blackwell | 8-bit | 256K | 77 tok/s | 0.5 s |
| M4 Max 128 GB | 8-bit | 256K | 25 tok/s | 9.9 s |
| M5 Max 128 GB | 8-bit | 256K | 27 tok/s | 2.9 s |
| Ryzen AI Max+ 395 128 GB | 8-bit | 256K | 14 tok/s | 5.7 s |
| DGX Spark 128 GB | 8-bit | 256K | 18 tok/s | 2.0 s |
| M3 Ultra 256 GB | 8-bit | 256K | 33 tok/s | 5.5 s |
| M5 Ultra 512 GB | 8-bit | 256K | 40 tok/s | 1.7 s |
| H100 SXM | 8-bit | 112K | 106 tok/s | 0.4 s |
| DGX Station GB300 | 8-bit | 256K | 138 tok/s | 0.2 s |
| 8× H200 NVL | 8-bit | 256K | 122 tok/s | 0.1 s |
Best fit is the highest quality, up to 8-bit, that leaves 16K of context. First reply is the wait for a 2K prompt. Speeds are rough, plus or minus 25%.
How to run it
- Ollama
ollama run nemotron-3.5-bolt:74b-a9b- llama.cpp
llama-server -hf unsloth/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-GGUF- vLLM
vllm serve nvidia/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-BF16
Questions
How much VRAM does Nemotron 3.5 Bolt 74B A9B need?
Nemotron 3.5 Bolt 74B A9B needs 42.3 GB for its weights at 4-bit, 43.3 GB with an 8K context and 43.5 GB with 32K. At 8-bit, 75.2 GB.
Can Nemotron 3.5 Bolt 74B A9B run on an RTX 4090?
Not on one card. The smallest machine here that runs it is the 2× RTX 3090.