How much VRAM does Serenity-26B-A4B need?

Q4_K_M is exactly 16,796,016,576 bytes (15.64 GiB) in weights. Add the KV cache, which depends on context length, plus roughly half a gigabyte of runtime overhead.

ReadyArt · text

Serenity-26B-A4B

ReadyArt/Serenity-26B-A4B

Serenity-26B-A4B at Q4_K_M is exactly 16,796,016,576 bytes (15.64 GiB / 16.80 GB) — an effective 5.062 bits per weight, not the nominal 4.

From the file· summed from 1 file(s)

Parameters

26.5B

Architecture

gemma4

Context

—

native (config.json)

License

apache-2.0

Shipped quantizations

● exact bytes, summed from published files

Quant	Size●	Exact bytes●	Effective bpw●	Tensors●	Publisher
IQ4_XS	13.10 GiB	14,063,809,472	4.239	—	ReadyArt
Q4_K_M	15.64 GiB	16,796,016,576	5.062	—	ReadyArt
Q5_K_M	17.82 GiB	19,132,891,072	5.766	—	ReadyArt
Q6_K2 shards	42.98 GiB	46,147,642,240	13.908	—	ReadyArt
Q8_02 shards	50.67 GiB	54,411,779,968	16.399	—	ReadyArt

Compare with

same modality, comparable size

vs Qwen3-Coder-30B-A3B-Instruct vs Qwen3.6-27B vs Qwen3.8-27B vs gemma-4-12B-it-qat-q4_0-unquantized vs gemma-4-E4B-it vs Qwen3-30B-A3B-Thinking-2507 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 13.91 GiB. The real file is 15.64 GiB, because a quantization is a mixture and some tensors are always kept at higher precision.

Architecture

Architecture unavailable — this repository is gated and no ungated mirror was found. Exact file sizes above are still authoritative; only the KV math needs the config.

Questions people ask

How much VRAM does Serenity-26B-A4B need?: Q4_K_M is exactly 16,796,016,576 bytes (15.64 GiB) in weights. Add the KV cache, which depends on your context length, plus roughly half a gigabyte of runtime overhead.
Which quantization of Serenity-26B-A4B 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.