localLLMriglocal, on your table

budget

Local LLM under $2,000

Under $2,000 (€1,787 incl. VAT) localLLMrig picks Qwen2.5 32B at Q4 as the largest model that fully fits somewhere. Global prices (USD), EUR available. Prices and tokens/s are estimates without a source.

This page is $2,000. The catalog compares against about €1,787 incl. VAT, an FX estimate.

Open in the calculator

Three setups

Qwen2.5 32B · Q4 · 8,192 ctx

Estimated memory

21.3 GBestimate

18.4 GB weights · 2.1 GB KV cache · 0.8 GB overhead

You didn't pick a model. Inside this budget, this is the largest model that fully fits somewhere.

  1. 01 · Fastest

    Used PC with RTX 3090 24GB

    Best single card here for 32B Q4. 70B Q4 does not fully fit.

    $1,433€1,280

    inside budget · used

    estimate

    tok/s

    24

    estimate

    needed

    21.3 GB

    full fit

    usable

    22.8 GB

    VRAM

    Why this one

    • The model fits entirely in VRAM or unified memory. No CPU offload.
    • It fits, with little room left. A longer context can push it over.
    • The indicative price is inside your budget.
    • High memory bandwidth, so a higher tokens-per-second estimate.
    • Second-hand or a previous generation. The price moves around.
    • Highest estimated speed in this shortlist.

    Parts

    • GeForce RTX 3090 24GB, used
    • Ryzen 7 5700X or similar
    • 64 GB DDR4
    • 850W PSU, decent airflow

  2. 02 · Other platform

    Mac mini M6 32GB

    32B Q4 fits, 70B does not. Apple lists 170 GB/s for the higher memory configuration.

    $1,398€1,249

    inside budget

    estimate

    tok/s

    3.9

    estimate

    needed

    21.3 GB

    full fit

    usable

    24.0 GB

    unified

    Why this one

    • The model fits entirely in VRAM or unified memory. No CPU offload.
    • It fits, with little room left. A longer context can push it over.
    • The indicative price is inside your budget.
    • Unified memory: weights and KV cache share one pool.
    • One box, little noise, no discrete GPU.
    • A different brand, so you don't see the same PC three times.

    Parts

    • Mac mini M6 with 32 GB (170 GB/s according to Apple)

  3. 03 · A step up

    PC with RTX 5090 32GB

    Highest bandwidth in a single card. 32 GB is plenty for 32B, still too tight for 70B Q4.

    $5,027€4,490

    over budget

    estimate

    tok/s

    43

    estimate

    needed

    21.3 GB

    full fit

    usable

    30.8 GB

    VRAM

    Why this one

    • The model fits entirely in VRAM or unified memory. No CPU offload.
    • Above your budget. Nothing fully fit inside the amount, or this is a faster step above it.
    • High memory bandwidth, so a higher tokens-per-second estimate.

    Parts

    • GeForce RTX 5090 32GB
    • Ryzen 7 9700X or similar
    • 96 GB DDR5
    • 1200W PSU, roomy case

With CPU offload, and therefore slower, this could also work: Mid-range PC with RTX 4060 Ti 16GB, ~4.2 tok/s, $1,668€1,490.

Tokens per second are estimated from memory bandwidth, not from a measured benchmark. Global prices (USD), EUR available. EUR includes VAT. USD is an FX estimate. No shop source. Run llama-bench yourself before you buy.

Want a build tailored to you? Custom report

Get new builds when GPU prices move

End of results

Common questions

What is the largest model that fully fits under $2,000?
In this dataset that is Qwen2.5 32B at Q4, with an estimated memory need of 21.3 GB at 8,192 tokens. That is an estimate. sourceUrl stays null.
Is a 70B model always included?
No. Llama 70B Q4 needs about 40 GB. At the lower amounts it does not fully fit. The page then shows a smaller model that does, or a fit above the budget when nothing inside it does.
Are the prices shop prices?
No. Every amount is indicative as of 14 October 2026 and has no shop source.
Local LLM under $2,000 · localLLMrig