family member

LFM2.5 24B-A2B

3.6 tokens a second on Linux · no GPU · 4 CPUs · 31 GB RAM. About $10 per million tokens here, against OpenRouter's $0.70: the API is 15× cheaper. Run it here for privacy or to keep work on your own machines, not to save money.

1 build of LFM2.5 24B-A2B, 1 signed run on real machines: different packagers, precisions and engines, each a different file with its own numbers.

3.6tok/s, fastest machine

$10per M tokens · $0.70 on an API

4Kcontext proved by recall

10.2 GBmemory, smallest build

1provider measured it

Run it on your gram Install a gram with one command. It works through the catalog on your own machine and signs every result. 1 provider has measured LFM2.5 24B-A2B so far.Use it in your own room Ask Evo in your gramx and it runs on a machine that proved it, at the metered $10 per million tokens. On your own machines it costs you nothing.Compare every build Put it beside every other model in one table: speed, price, context and the API price, one row each.

best on ollama · 1 build

trained 128K · proved 4K context, by unsloth IQ3_XXS

3.6 tok/s peak, by unsloth IQ3_XXS

What it fits on

The smallest build needs about 10.2 GB; 10 of 18 kinds of machine on the network can hold it. Bars: each kind's memory (its card's, else RAM) · ticks: every build, at the size it held plus a fifth for the prompt.

Linux, CPU only · 2 CPUs1 GB RAMnone
NVIDIA GeForce RTX 3050 Laptop GPU4 GB on the cardnone
NVIDIA GeForce GTX 16504 GB on the cardnone
AMD Radeon(TM) Vega 8 Mobile Graphics7 GB RAMnone
Windows, CPU only · 12 CPUs7 GB RAMnone
Apple silicon8 GB RAMnone
Apple M18 GB RAMnone
Mac, CPU only · 8 CPUs8 GB RAMnone
NVIDIA GeForce RTX 3080 Ti12 GB on the cardevery build
Tesla T415 GB on the cardevery build
Intel(R) UHD Graphics 73016 GB RAMevery build
Intel(R) Iris(R) Xe Graphics16 GB RAMevery build
Apple M416 GB RAMevery build
Apple M116 GB RAMevery build
NVIDIA L422 GB on the cardevery build
2× Tesla T429 GB on the cardevery build
Linux, CPU only · 4 CPUs31 GB RAMevery build
NVIDIA RTX PRO 6000 Blackwell Server Edition95 GB on the cardevery build

How far back it remembers, machine by machine

Filled: the needle came back at that prompt size · outlined: asked and lost · blank: not asked · trained for 128K. 0 walks ended because the model forgot, 1 because the machine could not hold the prompt.

machine · build	4K	why it stopped
Gram ·e61d3b 2× Tesla T4 · IQ3_XXS by unsloth		4K too slow to answer here

Carbon per million tokens

Lowest build on each machine · 0 of 1 read off a power sensor, the rest modelled from the parts · fewer grams is better

Gram ·e61d3b1.1 kg19.18 J/tok · modelled

Carbon is location-based: signed energy × the yearly average of the grid where the machine signed its place (EU grid average (applied to NL); Ember Global Electricity Review 2025 (2024 generation), location-based). Facility overhead is not included. A machine on its own solar is not credited until it can prove it.

Measured by

The people whose machines signed these numbers

Gram ·e61d3b1 run

proved by at least one build: Tool calling · Tool loop finished · Faithful to tool results · Chained tool calls · Structured output · Sustained output · Extended thinking · Writes code · Reasons about code · Synthesises sources · Follows instructions

Every build

Each its own file with its own page: the machines that ran it and their signatures.

build	engine	whose work	memory	context proved	best tok/s	machines
IQ3_XXS by unslothhf.co/unsloth/LFM2.5-24B-A2B-GGUF:IQ3_XXS	ollama	quantizer · unsloth	10.2 GB	4K	3.6	1

derivatives — somebody else's change to these weights

Fine-tunes, merges and abliterations of this member, each its own model with its own page. None of their numbers are counted above.

LFM2.5-24B-A2B-Uncensored-GGUFnot measured LFM2.5-24B-A2B-Turbo-Brilliance-50X-NEO-MAX-Imatrix-GGUFnot measured LFM2.5-24B-A2Bnot measured