[whatmodel.sh](../../README.md) / LFM2.5 24B-A2B # LFM2.5 24B-A2B Liquid AI · released 2026-10 · 24B params (2.45B active) · lfm1.0 **score 65.0** ±15.0 · #1 of 2 · from 7 benchmark results across 1 source · † no independently run results verified Score within the [GPQA Diamond](https://whatmodel.sh/?set=gpqa-diamond%40first-party-liquid-scores&score=set) comparison set only. ## What it's good at ● marks LFM2.5 24B-A2B among every model on that task or test, worst to best. 
task            vs the others            rank  counts  tests
agent tasks     █··········●         #1 of 2*     31%      4
coding          █··········●         #1 of 2*     10%      2
chat & writing  █··········●         #1 of 2*    8.9%      1
math                           not ranked yet       —      1
agent tasks · #1 of 2 · 31% of its score 
test                                result  vs the others in its run   difficulty        counts  why
WebArena-Verified (maker)            26.8%  ●··█···█·█·█ beats 0/4     hard · avg 32%       27%  sharp · at its level · small run
SWE-bench Verified (maker)           51.0%  ▁▁▁▁▁▁▂●█▆▁▁ beats 43/138  medium · avg 47%    4.1%  at its level · small run
BFCL V4 (December 2025 archive) (…   65.0%  ▁▁▁▂▆▆▅▆█▂●▃ beats 98/114                         —  not counted yet: no other model in the …
τ³-bench Banking (maker)              8.0%  ▂▂·▅█▂●▃▃▆▂▅ beats 11/24                          —  not counted yet: no other model in the …
coding · #1 of 2 · 10% of its score 
test                                result  vs the others in its run   difficulty        counts  why
SciCode (level unspecified) (make…   32.7%  ▁····▃▃▄●█▆▂ beats 16/40   hard · avg 30%      5.8%  at its level · small run
SWE-bench Verified (maker)           51.0%  ▁▁▁▁▁▁▂●█▆▁▁ beats 43/138  medium · avg 47%    4.1%  at its level · small run
chat & writing · #1 of 2 · 8.9% of its score 
test                                result  vs the others in its run    difficulty           counts  why
IFEval (aggregation unspecified) …   96.8%  ▁▂▄▂▂▄▆██▅▂● beats 145/145  very easy · avg 97%    8.9%  maxed out · small run
math · 0.0% of its score 
test              result  vs the others in its run  counts  why
MATH-500 (maker)   96.0%  ▁▃▂▃▄▇▅▇●▄▂▁ beats 60/85       —  not counted yet: no other model in the …
other tests (3) 
test                                result  vs the others in its run    difficulty            counts  why
Humanity's Last Exam (maker)          7.4%  ▁▁▁●▆▆▇██▆▁▃ beats 10/111   very hard · avg 7.0%    8.9%  small run
IFBench (aggregation unspecified)…   74.7%  ▂▆▂▄▄▄▂▄▄▆●▇ beats 65/86    easy · avg 74%          8.0%  small run
GPQA Diamond (maker)                 74.4%  ▁▁▁▁▂▄▅▅●██▄ beats 174/322  medium · avg 69%        0.8%  blunt · at its level · small run
## How the score was worked out Each result implies a score on its own, after adjusting for how hard that test is and who ran it. The model's score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its *counts* column above; here it is added up by who ran the tests. [Method](../method.md) 
source                                    results  share of the score
Liquid AI own-model benchmark facts (su…        7   64% ▇▇▇▇▇▇▇▇
prior: pull toward the middle                       36% ▇▇▇▇
weighted average                                   = 65.0
Not counted (3) 
source                  test                            score  why
Liquid AI own-model b…  BFCL V4 (December 2025 archiv…  65.0%  too few models have it yet
Liquid AI own-model b…  τ³-bench Banking                 8.0%  too few models have it yet
Liquid AI own-model b…  MATH-500                        96.0%  too few models have it yet
## How to run it ### llama.cpp (recommended) 1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md) ```sh # macOS / Linux curl -LsSf https://llama.app/install.sh | sh # Windows irm https://llama.app/install.ps1 | iex ``` 2. Run LFM2.5 24B-A2B (Q4_K_M, needs ~14.4 GB, from [huggingface.co/LiquidAI](https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-GGUF/blob/0912d33141f77e10705345a6732bfd83f0252fdc/README.md)) ```sh llama cli -hf LiquidAI/LFM2.5-24B-A2B-GGUF ``` 3. Or start a local server and open the address it prints (http://localhost:9931) ```sh llama serve -hf LiquidAI/LFM2.5-24B-A2B-GGUF ``` ### LM Studio 1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download) 2. Run LFM2.5 24B-A2B (Q4_K_M, needs ~14.4 GB, from [huggingface.co/LiquidAI](https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-GGUF/blob/0912d33141f77e10705345a6732bfd83f0252fdc/README.md)) ```sh lms get https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-GGUF -y ``` 3. Then run lms ls to see its name, and lms chat followed by that name ### Ollama 1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download) ```sh # Linux curl -fsSL https://ollama.com/install.sh | sh ``` 2. Run LFM2.5 24B-A2B (Q4_K_M, needs ~14.4 GB, from [huggingface.co/LiquidAI](https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-GGUF/blob/0912d33141f77e10705345a6732bfd83f0252fdc/README.md)) ```sh ollama run hf.co/LiquidAI/LFM2.5-24B-A2B-GGUF ``` Model repository and documentation: [huggingface.co/LiquidAI](https://huggingface.co/LiquidAI/LFM2.5-24B-A2B) [interactive version](https://whatmodel.sh/model/lfm2-5-24b-a2b) · [JSON](https://whatmodel.sh/api?format=json&all=1)