How much VRAM does DA3-BASE need?

Q4_K is exactly 1,406,986,144 bytes (1.31 GiB) in weights. Add the KV cache, which depends on context length, plus roughly half a gigabyte of runtime overhead.

depth-anything · text

DA3-BASE

depth-anything/DA3-BASE

DA3-BASE at Q4_K is exactly 1,406,986,144 bytes (1.31 GiB / 1.41 GB)

From the file· summed from 10 file(s)From the file· KV per layer

Parameters

—

Architecture

depthanything3

null layers

Context

—

native (config.json)

License

apache-2.0

Shipped quantizations

● exact bytes, summed from published files

Quant	Size●	Exact bytes●	Effective bpw●	Tensors●	Publisher
Q4_K10 shards	1.31 GiB	1,406,986,144	—	—	mudler
Q5_K9 shards	1.35 GiB	1,450,857,568	—	—	mudler
Q6_K9 shards	1.50 GiB	1,607,842,912	—	—	mudler
Q8_010 shards	1.90 GiB	2,042,823,328	—	—	mudler
F1610 shards	3.05 GiB	3,274,831,232	—	—	mudler
F3215 shards	13.94 GiB	14,969,376,768	—	—	mudler

Compare with

same modality, comparable size

vs Qwen3-4B vs Qwen3-8B vs HyperCLOVAX-SEED-Text-Instruct-1.5B vs Llama-3.2-1B-Instruct vs Qwen3-VL-8B-Instruct-abliterated-v1 vs Llama-3.1-8B-Instruct All alternatives →Will it fit on my card? →

Why other calculators give a different number

A parameters × bits ÷ 8 estimate puts Q4_K at roughly 0.00 GiB. The real file is 1.31 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

—

Hidden size

—

Vocab

—

Sliding window

none

SWA period

—

MLA

Experts

—

Experts per token

—

use_sliding_window

—

Questions people ask

How much VRAM does DA3-BASE need?: Q4_K is exactly 1,406,986,144 bytes (1.31 GiB) in weights. Add the KV cache, which depends on your context length, plus roughly half a gigabyte of runtime overhead.
Which quantization of DA3-BASE 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.