How much VRAM does AMD-OLMo-1B-SFT-DPO need?

Q4_K_M is exactly 733,520,640 bytes (0.68 GiB) in weights. Add the KV cache, which depends on context length, plus roughly half a gigabyte of runtime overhead.

How large is AMD-OLMo-1B-SFT-DPO's KV cache?

4.00 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.

amd · text

AMD-OLMo-1B-SFT-DPO

amd/AMD-OLMo-1B-SFT-DPO

AMD-OLMo-1B-SFT-DPO at Q4_K_M is exactly 733,520,640 bytes (0.68 GiB / 0.73 GB) — an effective 4.987 bits per weight, not the nominal 4. Its KV cache at 32K is 4.00 GiB.

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

Parameters

1.2B

Architecture

olmo

16 layers

Context

2,048

native (config.json)

License

apache-2.0

Shipped quantizations

● exact bytes, summed from published files

Quant	Size●	Exact bytes●	Effective bpw●	Tensors●	Publisher
IQ3_M	0.53 GiB	569,943,040	3.875	—	bartowski
Q3_K_L	0.61 GiB	651,469,568	4.429	—	lmstudio-community
Q3_K_L	0.61 GiB	651,469,824	4.429	—	bartowski
IQ4_XS	0.61 GiB	656,712,704	4.465	—	bartowski
Q4_0	0.64 GiB	692,364,288	4.707	—	bartowski
Q4_K_S	0.65 GiB	696,558,592	4.735	—	bartowski
Q4_K_M	0.68 GiB	733,520,640	4.987	—	lmstudio-community
Q4_K_M	0.68 GiB	733,520,896	4.987	—	bartowski
Q4_K_L	0.71 GiB	758,471,680	5.156	—	bartowski
Q5_K_S	0.77 GiB	824,484,864	5.605	—	bartowski
Q5_K_M	0.79 GiB	846,767,104	5.757	—	bartowski
Q5_K_L	0.81 GiB	871,717,888	5.926	—	bartowski
Q6_K	0.90 GiB	967,090,944	6.575	—	lmstudio-community
Q6_K	0.90 GiB	967,091,200	6.575	—	bartowski
Q6_K_L	0.92 GiB	992,041,984	6.744	—	bartowski
Q8_0	1.17 GiB	1,252,088,576	8.512	—	lmstudio-community
Q8_0	1.17 GiB	1,252,088,832	8.512	—	bartowski
F16	2.19 GiB	2,355,305,216	16.012	—	bartowski

KV cache by context

computed per layer

Context	KV cache (f16)●	Flat formula	Overstated by	Full / windowed / recurrent
4,096	0.50 GiB	0.50 GiB	—	16 / 0 / 0
8,192	1.00 GiB	1.00 GiB	—	16 / 0 / 0
16,384	2.00 GiB	2.00 GiB	—	16 / 0 / 0
32,768	4.00 GiB	4.00 GiB	—	16 / 0 / 0
65,536	8.00 GiB	8.00 GiB	—	16 / 0 / 0
131,072	16.00 GiB	16.00 GiB	—	16 / 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 0.62 GiB. The real file is 0.68 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

50,304

Sliding window

none

SWA period

—

MLA

Experts

—

Experts per token

—

use_sliding_window

—

Questions people ask

How much VRAM does AMD-OLMo-1B-SFT-DPO need?: Q4_K_M is exactly 733,520,640 bytes (0.68 GiB) in weights. Add the KV cache, which depends on your context length, plus roughly half a gigabyte of runtime overhead.
How large is AMD-OLMo-1B-SFT-DPO's KV cache?: 4.00 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 AMD-OLMo-1B-SFT-DPO 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.