How much VRAM does Lightning-1.7B need?

Q4_K_M is exactly 1,107,408,960 bytes (1.03 GiB) in weights. Add the KV cache, which depends on context length, plus roughly half a gigabyte of runtime overhead.

How large is Lightning-1.7B's KV cache?

3.50 GiB at 32K context with an f16 cache, computed per layer from the model's architecture. Quantizing the cache to q8_0 roughly halves it.

TitleOS · text

Lightning-1.7B

TitleOS/Lightning-1.7B

Lightning-1.7B at Q4_K_M is exactly 1,107,408,960 bytes (1.03 GiB / 1.11 GB) — an effective 5.149 bits per weight, not the nominal 4. Its KV cache at 32K is 3.50 GiB.

From the file· summed from 1 file(s)From the file· KV per layer

Parameters

1.7B

Architecture

qwen3

28 layers

Context

40,960

native (config.json)

License

mpl-2.0

Shipped quantizations

● exact bytes, summed from published files

Quant	Size●	Exact bytes●	Effective bpw●	Tensors●	Publisher
IQ2_M	0.65 GiB	695,181,376	3.232	—	bartowski
IQ3_XXS	0.70 GiB	754,360,384	3.507	—	bartowski
Q2_K	0.72 GiB	777,795,648	3.616	—	bartowski
IQ3_XS	0.78 GiB	834,222,144	3.879	—	bartowski
Q2_K_L	0.79 GiB	853,155,904	3.967	—	bartowski
Q3_K_S	0.81 GiB	867,252,288	4.032	—	bartowski
IQ3_M	0.83 GiB	895,662,144	4.165	—	bartowski
Q3_K_M	0.88 GiB	939,538,496	4.369	—	bartowski
Q3_K_L	0.93 GiB	1,003,501,632	4.666	—	bartowski
IQ4_XS	0.94 GiB	1,010,382,912	4.698	—	bartowski
IQ4_NL	0.98 GiB	1,054,423,104	4.903	—	bartowski
Q4_0	0.98 GiB	1,056,782,400	4.914	—	bartowski
Q4_K_S	0.99 GiB	1,060,190,272	4.929	—	bartowski
Q4_K_M	1.03 GiB	1,107,408,960	5.149	—	bartowski
Q4_1	1.06 GiB	1,142,503,488	5.312	—	bartowski
Q4_K_L	1.10 GiB	1,182,769,216	5.499	—	bartowski
Q5_K_S	1.15 GiB	1,230,583,872	5.722	—	bartowski
Q5_K_M	1.17 GiB	1,257,879,616	5.849	—	bartowski
Q5_K_L	1.24 GiB	1,333,239,872	6.199	—	bartowski
Q6_K	1.32 GiB	1,417,754,688	6.592	—	bartowski
Q6_K_L	1.39 GiB	1,493,114,944	6.942	—	bartowski
Q8_0	1.71 GiB	1,834,426,432	8.529	—	bartowski
BF16	3.21 GiB	3,447,349,024	16.029	—	bartowski

KV cache by context

computed per layer

Context	KV cache (f16)●	Flat formula	Overstated by	Full / windowed / recurrent
4,096	0.44 GiB	0.44 GiB	—	28 / 0 / 0
8,192	0.88 GiB	0.88 GiB	—	28 / 0 / 0
16,384	1.75 GiB	1.75 GiB	—	28 / 0 / 0
32,768	3.50 GiB	3.50 GiB	—	28 / 0 / 0
65,536	7.00 GiB	7.00 GiB	—	28 / 0 / 0
131,072	14.00 GiB	14.00 GiB	—	28 / 0 / 0

Compare with

same modality, comparable size

vs Qwen3-4B vs HyperCLOVAX-SEED-Text-Instruct-1.5B vs Llama-3.2-1B-Instruct vs ced-base vs gemma-3-1b-it vs Qwen3-1.7B All alternatives →Will it fit on my card? →

Why other calculators give a different number

A parameters × bits ÷ 8 estimate puts Q4_K_M at roughly 0.90 GiB. The real file is 1.03 GiB, because a quantization is a mixture and some tensors are always kept at higher precision.

Architecture

from config.json

Layers

Attention heads

KV heads

Head dim

128

Hidden size

2048

Vocab

151,936

Sliding window

none

SWA period

—

MLA

Experts

—

Experts per token

—

use_sliding_window

false

Questions people ask

How much VRAM does Lightning-1.7B need?: Q4_K_M is exactly 1,107,408,960 bytes (1.03 GiB) in weights. Add the KV cache, which depends on your context length, plus roughly half a gigabyte of runtime overhead.
How large is Lightning-1.7B's KV cache?: 3.50 GiB at 32K context with an f16 cache, computed per layer. Quantizing the cache to q8_0 roughly halves it, which is often the difference between a context length fitting and not.
Which quantization of Lightning-1.7B should I use?: Q4_K_M is the usual default. Pick the largest quantization that fits your card at the context you actually need — the table above gives exact sizes for every one published.