What hardware do you need to run Spark-X2.5 4B?

Spark-X2.5 4B at Q4_K_M is 2.6 GB of weights, with a context ceiling of 1024k tokens. Not yet rated: released after our last ratings pass. Dense 4B with a native million-token context and 200+ languages.

Cheapest machine that runs itMac mini M6, 16GB at $899

Fastest of the ones listedMac Studio M5 Ultra, 96GB — 233 tok/s at 32k context

Honest answer on costPays back in 75 years at 500k tokens a day.

How good is it, really?

It has not been placed on the intelligence index yet. See the whole table.

Summarising — not rated
Translation — not rated
Everyday coding — not rated
Reasoning & maths — not rated
Agentic work — not rated

What it costs either way

Nobody rents Spark-X2.5 4B by the token. The closest hosted match, Granite 4.2 8B, costs $0.06 per million input tokens and $0.25 per million output (OpenRouter, cheapest active endpoint, checked 2026-09-09). Buying a machine only beats that if you use it hard enough, for long enough, that the hardware price divides down below the rental bill.

Machines that run it

Machine	Price	Speed at 32k	Pay-back
Mac mini M6, 16GB	$899	30 tok/s estimated	Pays back in 75 years	Run the numbers
Strix Halo Framework Desktop, 32GB	$1,269	50 tok/s estimated	Pays back in 109 years	Run the numbers
MacBook Air M5 (13-inch), 16GB	$1,299	30 tok/s estimated	Pays back in 109 years	Run the numbers
MacBook Pro M5 (14-inch), 16GB	$1,999	30 tok/s estimated	Pays back in 167 years	Run the numbers
Mac Studio M5 Max, 36GB	$2,499	89 tok/s estimated	Pays back in 204 years	Run the numbers
DGX Spark GB10 Grace Blackwell, 128GB	$4,699	53 tok/s estimated	Pays back in 405 years	Run the numbers

One machine per family, cheapest first. Speeds are measured where a public benchmark exists and estimated from memory bandwidth otherwise; the calculator says which for any configuration.

The specifics

Parameters: 4.1B
Quantisation: Q4_K_M
Weights on disk: 2.6 GB
KV cache: 1.3 GB at 32k context — 3 sliding-window layers (512) per full-attention layer, which is what makes a million-token context tractable at this size.
Maximum context: 1024k tokens (1M native)
Licence: Apache 2.0
Sources: source 1, source 2