NVIDIA-Nemotron-3.5-Bolt-E4B-BF16 measured on a Mac Studio (M5 Max, 36GB)

Results

Measure	Result	Among measured models	Note
Reply time	about 2.6 s	20 of 39	Until one simple common-sense question is answered (median of 50)
Knowledge (high school to university)	76%	20 of 39	200 questions, ±6
Math (word problems)	93%	18 of 39	100 questions, ±7
Memory used	6.5GB	14 of 39
EN→JA translation	34 pts	4 of 35	out of 100 (chrF), about 0.6 s a sentence
JA→EN translation	56 pts	12 of 35	out of 100 (chrF), about 4.8 s a sentence

"Among measured models" is the place among the 39 models measured on this site (lower is better for time and memory).

Try it on your Mac

Install the Mac version of Ollama from ollama.com and start it.
Type this in Terminal. The first time, the model downloads; then you can chat with it.ollama run hf.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-4GB:Q4_K_M

It used 6.5GB here. To use it comfortably alongside other apps, a Mac with 16GB of memory or more is a good guide (the model then takes about half the memory).

About the model

Maker: NVIDIA
Size: 7.52B parameters
Quantization: Q4_K_M
Thinks before answering: No
Longest input at once: 65,536 tokens
Released (Hugging Face): Oct 5, 2026
Generation speed: 86.9 tok/s
Start-up (loading): about 1.1 s
Measured on: Oct 24, 2026
Ollama: 0.34.3
Hugging Face: nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16
Series: Nemotron

What the parts of a name mean (size, quantization and so on): how to read a model name.

Models of similar memory

Qwen3.5-9B (6.5GB of memory, knowledge 84%, reply about 40.3 s)
DeepSeek-R1-0528-Qwen3-8B (6.4GB of memory, knowledge 72%, reply about 9.4 s)
Ornith-1.5-9B (6.4GB of memory, knowledge 77%, reply about 2.5 s)
GLM-4-9B-0414 (6.7GB of memory, knowledge 72%, reply about 0.1 s)
Ornith-1.0-9B (6.2GB of memory, knowledge 85%, reply about 3.8 s)

Compare the main models on the text page. The series page lists every model measured.