NVIDIA Nemotron 3.5 Bolt E4B: VRAM requirements by quant
Key facts
NVIDIA Nemotron 3.5 Bolt E4B needs about 7.3 GB at standard quality (Q4, estimate).
Checked . NVIDIA Q4_K_M GGUF (official).
NVIDIA Nemotron 3.5 Bolt E4B is a small, fast AI model you can download and run on your own computer. It needs about 7.3 GB of memory at standard quality.
LIVE IN BROWSER · $0 · NO LOGIN
Test comparable open models (Recursive Flint 40B vs Gemma 4 31B) free in your browser
Open full 8-model playground ↗Click a starter prompt below or type your own question to watch Recursive Flint stream side by side against Gemma 4 31B right on this page:
Memory needed
About 7.3 GB
At standard quality, with room for a conversation of about 12 pages.
Get a 1-line alert when a faster quant for NVIDIA Nemotron 3.5 Bolt E4B drops — or when GPUs that run it drop >10%
Get a 1-line alert when a new open-weight model fits 8 GB VRAM (7.3 GB at Q4 for NVIDIA Nemotron 3.5 Bolt E4B) or when used RTX 3090 / 4090 & Mac Studio prices drop >10% — weekly, zero spam.
Machines that run NVIDIA Nemotron 3.5 Bolt E4B
Each of these holds the whole model in memory. Speed is in tokens per second (tok/s); from about 8 it keeps up with your reading.
EU shops · Shown to visitors in Europe.US shops · Shown to visitors outside Europe.
If you buy through these links we may earn a commission. It doesn't change our ranking. How shop links work
Mac mini M4 16GB
16 GB shared memory · second-hand
Comfortable~8.9 tok/s$615€549usedunverified
Machine details →Mac mini M6 16GB
16 GB shared memory
Comfortable~11.4 tok/s$989€1,069new
Machine details →Used PC with RTX 3090 24GB
24 GB graphics memory · second-hand
Fast~111 tok/s$2,413€2,346estimateused parts
Machine details →The buttons look up the graphics card, the part of this PC worth comparing.
PC with Radeon RX 7900 XTX 24GB
24 GB graphics memory · second-hand
Fast~109 tok/s$2,422€2,354estimateused parts
Machine details →The buttons look up the graphics card, the part of this PC worth comparing.
PC with RTX 5090 32GB
32 GB graphics memory
Fast~208 tok/s$7,834€8,467estimate
Machine details →The buttons look up the graphics card, the part of this PC worth comparing.
Where our price is for a used card or Mac, these shops mostly sell new ones, so their price can be higher.
As an Amazon Associate I earn from qualifying purchases.
Als Amazon-Partner verdiene ich an qualifizierten Verkäufen.
Als Amazon-Partner verdien ik aan in aanmerking komende aankopen.
About these prices
Each price is the amount we last wrote down for that machine, with its date on the machine page. "Estimate" is our own figure: for a complete PC we add our estimate for the other parts to the graphics card price, and some Mac prices are estimates, not a checked shop listing. "Used" is a second-hand price. "Unverified" means we have no checked shop listing for that exact amount. The shop buttons open a search for that machine at each shop, so you see today's price there. Prices excl. shipping unless marked. Euro prices incl. VAT where the seller charges it. Dollar prices are converted from our euro prices without the 21% EU VAT, before US sales tax.
Or rent one by the hour: RTX A6000 at Vast.ai holds it, for $0.40 an hour (median of live offers, 21 Oct 2026). Compare GPU cloud prices
Speeds and memory are our estimate from the hardware specs, not a measured test. Prices are dated and may change. How we calculate
Memory at each quality level
A lower quality level makes the model smaller, so it fits on a smaller machine. Standard (Q4) is what most people run.
| Quality | Memory needed |
|---|---|
| Q4 · standard (smallest that keeps quality) | 7.3 GB |
| Q5 · a bit sharper, a bit bigger | 8.4 GB |
| Q8 · near full quality, almost twice Q4 | 9.6 GB |
| FP16 · full quality, over three times Q4 | 17.1 GB |
With a conversation of 8,192 tokens · about 12 pages of text.
Sources and method
At standard quality: 6.2 GB for the model itself (a measured file), 0.3 GB for a conversation of 8,192 tokens, and 0.8 GB for the software.
Size source: huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-4GB. The published size is rounded, so ours is an estimate.
NVIDIA's small Nemotron 3.5: 8B parameters. The card calls 4B of that effective, from per-layer embeddings, not a mixture of experts. Open weights under OpenMDW-1.1; the Gemma 4 base is Apache 2.0. The card says text in and text out, and context up to 64K (65,536). The GGUF header and Ollama list 128K. The config, Ollama and the unsloth GGUFs also include an image encoder; the mmproj files (0.56–0.99 GB) are not in these totals. Ollama's :e4b download was 7.91 GB on 23 Oct 2026 (the 6.18 GB Q4_K_M plus a projector and a draft). Without :e4b, Ollama pulls the 74B model. 18 of 42 layers share key and value state, which this estimate does not model, so the conversation figure is conservative.
Licence
Commercial use: yes
Commercial use allowed. Keep the licence and copyright notice when you share the weights.
Licence: OpenMDW-1.1 · checked 24 Oct 2026
Our reading of the licence, not legal advice.
Will NVIDIA Nemotron 3.5 Bolt E4B run on my graphics card?
Builder interview
Fabio Matricardi (@fabiomatricardi) runs NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q5_K_XL.gguf in llama.cpp Vulkan on a Lenovo ThinkPad with an AMD Radeon 780M iGPU (32 GB unified RAM) and on a Dell i5 workplace laptop with an Intel Ultra iGPU (32 GB unified RAM). For opencode agent-harness runs he stays at Q6, with Q5_K_M as an occasional exception, so structured output and tool calls do not drift. Self-reported, 25 October 2026. Read Fabio Matricardi's interview.
Common questions
- How much memory does NVIDIA Nemotron 3.5 Bolt E4B need?
- About 7.3 GB at standard quality (Q4) with a conversation of 8,192 tokens: 6.2 GB for the model itself, 0.3 GB for the conversation and 0.8 GB for the software. The model size is a measured file; the conversation and software parts are our estimate.
- Does NVIDIA Nemotron 3.5 Bolt E4B fit on a single 24 GB graphics card?
- Yes, by our estimate. It needs about 7.3 GB, and we count 23.4 GB of a 24 GB card as usable.
- Are the speeds measured?
- No. They are our estimate from how fast each machine's memory is. Real speed also depends on the software and its settings.
- Can I run NVIDIA Nemotron 3.5 Bolt E4B on an 8GB or 12GB GPU?
- Yes, by our estimate. At Q4_K_M (6.18 GB measured GGUF + 0.27 GB conversation memory at 8,192 tokens + 0.80 GB working space = 7.25 GB / 7.3 GB total estimate), it fits a 12 GB GPU (11.4 GB usable; cheapest on our list: GeForce RTX 3060 12GB at €261 / $292, about 43 tok/s estimate) and just fits the 7.4 GB an 8 GB card would leave usable under the same 0.6 GB reserve (this list has no 8 GB card). For more headroom on 8 GB, UD-Q5_K_XL (5.83 GB measured GGUF, 6.9 GB total estimate) or UD-Q4_K_XL (5.06 GB measured GGUF, 6.1 GB total estimate) or UD-Q3_K_XL (4.22 GB measured GGUF, 5.3 GB total estimate) or UD-IQ2_M (3.53 GB measured GGUF, 4.6 GB total estimate). The model notes put the mmproj image files at 0.56–0.99 GB, and those files are not included in these totals.
- Can I run NVIDIA Nemotron 3.5 Bolt E4B on 24GB?
- Yes, by our estimate. At Q4_K_M (6.18 GB measured GGUF + 0.27 GB conversation memory at 8,192 tokens + 0.80 GB working space = 7.25 GB / 7.3 GB total estimate), NVIDIA Nemotron 3.5 Bolt E4B fits inside the 23.4 GB usable memory of a 24 GB card. The cheapest 24 GB card on our list is GeForce RTX 3090 24GB at €1,246 / $1,395. A 24 GB card also holds BF16 (15.05 GB measured GGUF, 16.1 GB total estimate) and UD-Q8_K_XL (8.19 GB measured GGUF, 9.3 GB total estimate) and Q8_0 (8.03 GB measured GGUF, 9.1 GB total estimate) and UD-Q6_K_XL (7.26 GB measured GGUF, 8.3 GB total estimate).
- Can I run NVIDIA Nemotron 3.5 Bolt E4B on a 16GB or 128GB Mac?
- Yes, by our estimate. It fits on Mac mini M4 16GB (10.7 GB usable of 16 GB, €549 / $615, price unverified, about 8.9 tok/s estimate at Q4_K_M). MLX-4bit (4.2 GB measured MLX, 5.3 GB total estimate) is the Apple Silicon build. It also fits Mac Studio M4 Max 128GB and Mac Studio M5 Max 128GB (96.0 GB usable of 128 GB).
- How fast is NVIDIA Nemotron 3.5 Bolt E4B on an RTX 5090, RTX 4090, and RTX 3090?
- Our hardware-spec estimates at Q4 are about 208 tok/s on an RTX 5090, about 119 tok/s on an RTX 4090, about 111 tok/s on an RTX 3090, about 109 tok/s on an RX 7900 XTX, about 43 tok/s on an RTX 3060 12GB (estimates, not measured benchmark runs).
- Why does Ollama download 7.91 GB or pull a 74B model for Nemotron 3.5 Bolt?
- Include the :e4b tag (ollama run nemotron-3.5-bolt:e4b). Without :e4b, Ollama pulls the larger model (NVIDIA Nemotron 3.5 Bolt 74B-A9B, https://localllmrig.com/can-i-run/nemotron-3-5-bolt-74b). The model notes record that :e4b download as 7.91 GB, bundling the 6.18 GB Q4_K_M weights with a projector and a draft model.
- Can I test Nemotron 3.5 models live in my browser before downloading?
- Yes. At https://localllmrig.com/ and in the inline test box on https://localllmrig.com/can-i-run/nemotron-3-5-bolt-e4b, you can test 10 open models for free with no account, including NVIDIA Nemotron 3.5 Lightning (https://localllmrig.com/can-i-run/nemotron-3.5-lightning), Nemotron 3 Nano Omni, NVIDIA Nemotron 3.5 Super VL (https://localllmrig.com/can-i-run/nemotron-3.5-super-vl).
- Can I try NVIDIA Nemotron 3.5 Bolt E4B online for free?
- Yes. NVIDIA Nemotron 3.5 Bolt E4B is on the free chat on the home page. Pick it there, ask a question, and read the answer in your browser. No login and no install.
- Do I need a GPU to test it?
- No. The free chat runs in your browser. It does not download the model and it does not use a graphics card on your computer. A graphics card only matters if you run the model yourself.
- Do I need an account to try NVIDIA Nemotron 3.5 Bolt E4B?
- No. The free chat needs no account and no login.