What hardware do you need to run LFM2.5 24B-A2B?

LFM2.5 24B-A2B at Q4_K_M is 15 GB of weights, with a context ceiling of 128k tokens. Liquid's largest LFM2.5: 24B total with 2.45B active, post-trained for agents and tool use. It reasons before answering, so a reply costs more output tokens than its size suggests. On the cheapest machine that runs it, a Mac mini M6, 24GB, it generates 24 tok/s at 32k of context, worked out from memory bandwidth.

How good is LFM2.5 24B-A2B, really?

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

What LFM2.5 24B-A2B costs either way

Renting the same model costs $0.05 per million input tokens and $0.7 per million output (OpenRouter, cheapest active endpoint, checked 2026-10-20). 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. What a million tokens costs each way puts the two prices side by side.

Machines that run LFM2.5 24B-A2B

Every machine here runs it, but not to the same length. The Mac mini M6, 24GB stops at 64k; the Strix Halo Framework Desktop, 32GB takes it to 128k, this model's own ceiling. The weights are the same size on every machine; what differs is the memory left for the key-value cache, which grows with every token you keep.

The table is the cheapest machine in each family, not the only one that runs it: 49 of the 56 machines on this site hold it at 32k. Within a family only memory decides, so that is every Mac Studio and Strix Halo box at any size, every Mac mini, MacBook Pro and NVIDIA card with 24GB or more, the Radeon AI PRO R9700, 32GB and the DGX Spark, 128GB. One table sets every machine against the others: price, the memory its GPU can use, what it holds and how long it takes to pay for itself.

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 longest context is the longest setting the calculator offers that the machine still holds this model at, cache included, and no machine is shown taking it past its own 128k limit. Each one opens the calculator on that machine at that length.

For scale, the calculator starts from 80 tok/s for a hosted API and times a local machine against it. 3 of the 7 above reach it, and the slowest is 21 tok/s.

The cheapest machine that runs it is sold with 24 GB. There is a page on what 24 GB runs, machine by machine: it names the 7 machines here sold at that size, and the 25 current models the roomiest of them holds at 32k, this one among them.

Run LFM2.5 24B-A2B at 128k on the Strix Halo Framework Desktop, 32GB

How much memory does LFM2.5 24B-A2B need at 128k?

Every figure above is priced at 32k of context. At 128k, this model's own ceiling, LFM2.5 24B-A2B needs 17.2 GB rather than 15.2 GB. The weights are the same either way; the key-value cache goes from 0.7 GB to 2.7 GB.

33 of the 37 machines here still sold at a published price hold LFM2.5 24B-A2B at 32k and 30 hold it at 128k. The cheapest still sold goes from the Mac mini M6, 24GB at $1,099 to the Framework Desktop, 32GB at $1,269.

On the Framework Desktop, 32GB it runs 65 tok/s estimated at 32k and 34 tok/s estimated at 128k. Every token generated reads the whole cache back, so a longer window is slower as well as bigger.

The specifics

Parameters

24B, of which 2.5B are active per token

Quantisation

Q4_K_M

Weights on disk

15 GB

KV cache

0.7 GB at 32k context — Only 10 of 40 layers are attention; the rest are short convolutions that keep a fixed-size state. head_dim derived from hidden_size ÷ heads, as the config omits it. How weights and cache add up.

Maximum context

128k tokens

Licence

LFM 1.0 (see repo)

Sources

LiquidAI/LFM2.5-24B-A2B, LiquidAI/LFM2.5-24B-A2B-GGUF