Skip to content

Nemotron 3.5 Bolt 74B A9B on 4× RTX 3090: VRAM, speed and fit

Nemotron 3.5 Bolt 74B A9B at 8-bit runs on the 4× RTX 3090 with 256K of context at about 50 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on 4× RTX 3090 · 8-bit (Q8_0) · Ollama / LM Studio · Chat

GPU memory
79.0 GB
of 96.0 GB
Context
256K
auto
Speed
50 tok/s
Very fast
First reply
0.5 s
2K prompt

Where the memory goes

Model
74.1 GB
Context
2.2 GB
Overhead
2.7 GB
OS
0.0 GB
Free
17.0 GB
Memory
79.0 GB of 96.0 GB
Context
256K tokens · auto
Speed
50 tok/s · very fast
First reply
0.5 s for a 2K prompt
Running cost
$20/mo at home · $122/mo on a cloud 4× RTX A4500

Validation: Validated

What Nemotron 3.5 Bolt 74B A9B on 4× RTX 3090 was checked against
CheckTargetResultStatus
Fits in GPU memory96.0 GB usable79.0 GB neededPass
Headroom5% free17.0 GB free (18%)Pass
Context for chat16K or more256KPass
Writing speed15 tok/s or more50 tok/sPass
First replyunder 3 s, 2K prompt0.5 sPass

Targets: a reply you can read along with, a first reply the use case tolerates, and 5% of memory left free. Calculated from published specs, not benchmarked.

The setup

Model
Nemotron 3.5 Bolt 74B A9B
Parameters
74.9B, 9B active
Precision
8-bit (Q8_0)
Context
256K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
4× RTX 3090
GPU can use
96.0 GB of 96.0 GB

Running cost

At home
$20/mo
920 W at the wall
Renting instead
$122/mo
4× RTX A4500 at $1.00/h

4 h a day, power at $0.18 per kWh, cloud at RunPod on-demand prices as of Nov 21, 2026.

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

Nemotron 3.5 Bolt 74B A9B on similar machines

All machines: Nemotron 3.5 Bolt 74B A9B VRAM requirements. Everything that runs here: what can the 4× RTX 3090 run?

Other models that run on 4× RTX 3090

Questions

Can Nemotron 3.5 Bolt 74B A9B run on 4× RTX 3090?

Nemotron 3.5 Bolt 74B A9B at 8-bit runs on the 4× RTX 3090 with 256K of context at about 50 tokens a second.

Report VR-15X6U66, calculated Nov 22, 2026 from published specs, not benchmarked.