Run BGE-M3 on a VPS.

Prices checked

The cheapest VPS that runs BGE-M3 is OVHcloud VPS-1 2027 at $5.02 a month, with 4 GB of RAM for the 2.6 GB it needs, checked .

Best plans for BGE-M3

Cheapest first, among plans that fit.

  1. 1

    VPS-1 2027 · 2 vCPU / 4 GB · EU

    OVHcloud · 2C · 4 GB · 40 GB NVMe

    $5.02/mo
  2. 2

    VPS 1 · 4 vCPU / 6 GB · Europe

    Leaseweb · 4C · 6 GB · 100 GB NVMe

    $5.02/mo
    View
  3. 3

    VPS-1 2027 · 2 vCPU / 4 GB · US

    OVHcloud · 2C · 4 GB · 40 GB NVMe

    $5.35/mo
    View
  4. 4

    VPS 1 · 4 vCPU / 6 GB · UK

    Leaseweb · 4C · 6 GB · 100 GB NVMe

    $6.02/mo
    View
  5. 5

    CX23

    Hetzner · 2C · 4 GB · 40 GB NVMe

    not available now
    $6.13/mo
    View
  6. 6

    VPS-1 LZ 2027 · 2 vCPU / 4 GB · EU

    OVHcloud · 2C · 4 GB · 40 GB NVMe

    $6.13/mo
    View

What it needs

RAM needed

2.6 GB

CPU inference at native format

Weights

1.1 GB

native format

OS and runtime

1.5 GB

headroom

Multilingual dense + sparse embeddings for hybrid search.

Size
0.568B parameters
Context
8K tokens
Kind
embedding
Released
2024-01
License
MIT

Facts from Hugging Face: sizes from bits per weight, 29,675,435 downloads.

ollama run bge-m3model card various

Or rent a GPU

every GPU price

1 provider rents the RTX A5000, with 24 GB of GPU memory against the 2.6 GB BGE-M3 needs.

Cheapest board that fits

RTX A5000

24 GB of GPU memory

On demand

$0.16per GPU an hour

RunPod

Held for a month

$116.80

730 hours of RTX A5000 · Community Cloud

How we count

On a CPU-only VPS BGE-M3 needs about 2.6 GB of RAM: 1.1 GB of weights in the model's native format and 1.5 GB for the OS and runtime. 308 of the 527 plans in the index fit and include a disk. A plan runs it comfortably with 20% headroom over the total, and is a tight fit with exactly enough.

A GPU rate is per GPU for an hour, and the monthly figure is 730 hours of the whole instance at it, tax excluded.

Frequently asked

How much RAM does BGE-M3 need?
About 2.6 GB for CPU inference at Q4_K_M: 1.1 GB of weights, 0 MB of KV cache at 8,192 tokens of context, and 1.5 GB of headroom. Longer contexts need more KV cache (0 KB per token for this model).
What is the cheapest VPS that can run BGE-M3?
OVHcloud VPS-1 2027 · 2 vCPU / 4 GB · EU (4 GB RAM, 2 vCPU) at $5.02/mo excl. VAT runs it comfortably as of 31 Oct 2026.
How fast will BGE-M3 run on a VPS without a GPU?
Embedding models are small and run well on CPUs. Throughput scales with vCPUs rather than RAM.

Same lineup