[whatmodel.sh](../../README.md) / LFM2 24B-A2B
# LFM2 24B-A2B
Liquid AI · released 2026-02 · 23.8B params (2.3B active) · lfm1.0
**score 51.2** ±1.1 · #344 of 732 · from 11 benchmark results across 1 source
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks LFM2 24B-A2B among every model on that task or test, worst to best.
task vs the others rank counts tests
math ▁▁▃▅█●▄█▄▂▁▁ #222 of 431 31% 4
math · #222 of 431 · 31% of its score
Beats [MiniMax-M2.1](minimax-m2-1.md), [DeepSeek V3 (Dec '24)](deepseek-v3.md), [GPT-4o (2024-08-06)](gpt-4o-2024-08-06.md) here, though they rank higher overall. Trails [Phi-2 (2.7B)](phi-2.md), [Granite 8B Code Base (4K)](granite-8b-code-base-4k.md), [Mixtral 8x7B v0.1 (base)](mixtral-8x7b-v0-1.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
GSM8K 87.2% █····█··●·▄█ beats 4/7 very easy · avg 86% 14% sharp
MATH (subset unspecified) 83.9% █▄··▄●▄····▄ beats 5/7 easy · avg 80% 7.5% typical
PolyMath (Marco evaluation) 29.3% ▄··▄█●··▄··▄ beats 5/7 hard · avg 29% 5.2% at its level
MGSM (Marco evaluation) 67.8% ▄·▄·▄●▄·▄··█ beats 3/7 easy · avg 69% 4.7% at its level
other tests (7)
test result vs the others in its run difficulty counts why
GPQA Diamond 39.7% ▄···█·▄·●·▄▄ beats 5/7 hard · avg 37% 20% sharp
MMLU 74.9% ▄···▄▄·●█··█ beats 3/7 easy · avg 74% 15% at its level
MMLU-Pro 57.6% ▄····█·▄●·▄▄ beats 5/7 medium · avg 54% 11% at its level
MMLU-ProX-Lite (Marco) 43.3% █··▄··●··█·▄ beats 4/7 medium · avg 42% 8.0% at its level
GlobalMMLU (Marco evaluation) 57.0% ▃··▃▃··●···█ beats 4/7 medium · avg 56% 5.8% at its level
MMMLU (Marco evaluation) 62.3% ▄···▄·█·●·▄█ beats 4/7 medium · avg 59% 5.0% at its level
AGIEval (Marco suite unspecified) 49.0% ▄··▄█▄··●··█ beats 5/7 medium · avg 46% 4.0% blunt · at its level
source results share of the score
Marco-MoE: Open Multilingual Mixture-of… 11 99% ▇▇▇▇▇▇▇▇▇▇▇▇
prior: pull toward the middle 1% ▇
weighted average = 51.2
Licences: Marco-MoE: Open Multilingual Mixture-of-Expert Language Models with Efficient Upcycling ([CC BY 4.0](../../licenses/cards-final-NOTICE.txt))
## 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 24B-A2B (Q4_K_M, needs ~14.4 GB, from [huggingface.co/LiquidAI](https://huggingface.co/LiquidAI/LFM2-24B-A2B-GGUF/tree/6247d03b87c27c05a258feb5acdfb4d5efbf0bd8))
```sh
llama cli -hf LiquidAI/LFM2-24B-A2B-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf LiquidAI/LFM2-24B-A2B-GGUF:Q4_K_M
```
Other versions: Q8_0 (~25.4 GB) `llama cli -hf LiquidAI/LFM2-24B-A2B-GGUF:Q8_0`
### 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 24B-A2B (Q4_K_M, needs ~14.4 GB, from [huggingface.co/LiquidAI](https://huggingface.co/LiquidAI/LFM2-24B-A2B-GGUF/tree/6247d03b87c27c05a258feb5acdfb4d5efbf0bd8))
```sh
lms get https://huggingface.co/LiquidAI/LFM2-24B-A2B-GGUF@Q4_K_M -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~25.4 GB) `lms get https://huggingface.co/LiquidAI/LFM2-24B-A2B-GGUF@Q8_0 -y` · 4bit (~13.4 GB, Mac only) `lms get https://huggingface.co/LiquidAI/LFM2-24B-A2B-MLX-4bit -y` · 8bit (~25.3 GB, Mac only) `lms get https://huggingface.co/LiquidAI/LFM2-24B-A2B-MLX-8bit -y`
### 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 24B-A2B (Q4_K_M, needs ~14.78 GB, from [ollama.com](https://ollama.com/library/lfm2/tags))
```sh
ollama run lfm2:24b
```
Other versions: Q8_0 (~29.06 GB) `ollama run lfm2:24b-q8_0`
### MLX (Mac)
1. Install MLX (Mac). Apple silicon Macs only, text only.
```sh
# with uv
uv tool install mlx-lm
# or
pip install mlx-lm
```
2. Run LFM2 24B-A2B (4bit, needs ~13.4 GB, Mac only, from [huggingface.co/LiquidAI](https://huggingface.co/LiquidAI/LFM2-24B-A2B-MLX-4bit/tree/fb67c8c23d38cd4d7a9a6415ab80eefe83feecae))
```sh
mlx_lm.chat --model LiquidAI/LFM2-24B-A2B-MLX-4bit
```
3. Or start an OpenAI-compatible server at http://localhost:8080/v1
```sh
mlx_lm.server --model LiquidAI/LFM2-24B-A2B-MLX-4bit
```
Other versions: 8bit (~25.3 GB, Mac only) `mlx_lm.chat --model LiquidAI/LFM2-24B-A2B-MLX-8bit`
Model repository and documentation: [huggingface.co/LiquidAI](https://huggingface.co/LiquidAI/LFM2-24B-A2B)
[interactive version](https://whatmodel.sh/model/lfm2-24b-a2b) · [JSON](https://whatmodel.sh/api?format=json&all=1)