[whatmodel.sh](../../README.md) / Gemma 3n E2B IT # Gemma 3n E2B IT Google · released 2025-06 · 5.44B params (2B active) · gemma **score 44.7** ±1.5 · #482 of 732 · from 12 benchmark results across 4 sources Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at ● marks Gemma 3n E2B IT among every model on that task or test, worst to best. 
task    vs the others      rank  counts  tests
coding  ●··········█    #2 of 2     25%      3
math    ●··········█   #2 of 2*     28%      2
coding · #2 of 2 · 25% of its score 
test                                result  vs the others in its run    difficulty      counts  why
HumanEval (maker)                    66.5%  ▁▁▁▁▂▂▆▆●▅▂▁ beats 127/205  easy · avg 78%     14%  sharp · at its level · small run
MBPP (maker)                         56.6%  ▁▁▁▄▅█●▅▄▁▂▁ beats 51/86    easy · avg 68%    6.9%  at its level · small run
LiveCodeBench v5 (window unspecif…   18.6%  ●▃·····▃█··▅ beats 0/7      hard · avg 32%    4.6%  small run
math · #2 of 2 · 28% of its score 
test               result  vs the others in its run  difficulty      counts  why
MGSM (maker)        53.1%  ▂·▁▁▁▁▁▂▅●▃▂ beats 26/42  easy · avg 67%     15%  sharp · at its level · small run
AIME 2025 (maker)    6.7%  ▁·▁●▁▂▅▆█▇▅▁ beats 5/133  hard · avg 17%     13%  sharp · small run
other tests (9) 
test                                result  vs the others in its run   difficulty        counts  why
Open Persian LLM Leaderboard v2 a…   52.0%  ▂▄▂▃█●▃▆▃▅▂▂ beats 36/81   medium · avg 57%     10%  blunt · at its level
MMLU (maker)                         60.1%  ▁▁▁▁▁▂▂▅●█▆▄ beats 88/214  medium · avg 70%     10%  at its level · small run
MMLU-Pro (maker)                     40.5%  ▁▁·▁▁▁●▂▄▅█▁ beats 27/188  medium · avg 53%    7.9%  at its level · small run
Open PT-LLM Leaderboard average (…   56.3%  ▃▁▁▁▁▁▁▁●█▅▁ beats 80/276  medium · avg 72%    5.3%  at its level · noisy
Global-MMLU (maker)                  55.1%  ●▅▃·▃···▅·▅█ beats 0/11    medium · avg 63%    4.5%  at its level · small run
EuroEval HellaSwag (English), tes…   66.4%  ▃·▁▁▂▁▂▅●▇▄▁ beats 42/100  medium · avg 67%    4.2%  sharp · at its level · noisy
EuroEval Life in the UK (English …   79.4%  ▁▁▁▁▁▁▁▂▄●▄▃ beats 63/146  easy · avg 82%      3.3%  at its level · noisy
GPQA Diamond (Google RelaxedAccur…   24.8%  ······●····· beats 1/1                            —  not counted yet: no other model in the …
Codegolf v2.2 (maker)                11.0%  ●··········█ beats 0/1                            —  not counted yet: no other model in the …
## 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
Google DeepMind Gemma 3n E2B IT model c…        8   76% ▇▇▇▇▇▇▇▇▇
Open Persian LLM Leaderboard v2                 1   10% ▇
EuroEval leaderboards                           2    8% ▇
Open Portuguese LLM Leaderboard                 1    5% ▇
prior: pull toward the middle                        1% ▇
weighted average                                   = 44.7
Licences: Google DeepMind Gemma 3n E2B IT model card ([CC-BY-4.0](../../licenses/cards-final-NOTICE.txt)) · Open Persian LLM Leaderboard v2 ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt)) · EuroEval leaderboards ([MIT](https://github.com/EuroEval/leaderboards/blob/1a77f1d2f8ba1ba05ab73287fb090d76bf41b2e0/LICENSE)) · Open Portuguese LLM Leaderboard ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt))
Not counted (2) 
source                  test                            score  why
Google DeepMind Gemma…  GPQA Diamond (Google RelaxedA…  24.8%  too few models have it yet
Google DeepMind Gemma…  Codegolf v2.2                   11.0%  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 Gemma 3n E2B IT (Q4_K_M, needs ~3 GB, from [huggingface.co/unsloth](https://huggingface.co/unsloth/gemma-3n-E2B-it-GGUF/tree/d7bd20d510eff26f3f95260c50f032fee8931a35)) ```sh llama cli -hf unsloth/gemma-3n-E2B-it-GGUF:Q4_K_M ``` 3. Or start a local server and open the address it prints (http://localhost:9931) ```sh llama serve -hf unsloth/gemma-3n-E2B-it-GGUF:Q4_K_M ``` Other versions: Q8_0 (~4.8 GB) `llama cli -hf unsloth/gemma-3n-E2B-it-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 Gemma 3n E2B IT (Q4_K_M, needs ~3 GB, from [huggingface.co/unsloth](https://huggingface.co/unsloth/gemma-3n-E2B-it-GGUF/tree/d7bd20d510eff26f3f95260c50f032fee8931a35)) ```sh lms get https://huggingface.co/unsloth/gemma-3n-E2B-it-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 (~4.8 GB) `lms get https://huggingface.co/unsloth/gemma-3n-E2B-it-GGUF@Q8_0 -y` · 4bit (~4.5 GB, Mac only) `lms get https://huggingface.co/lmstudio-community/gemma-3n-E2B-it-MLX-4bit -y` · 8bit (~6.7 GB, Mac only) `lms get https://huggingface.co/lmstudio-community/gemma-3n-E2B-it-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 Gemma 3n E2B IT (Q4_K_M, needs ~3.76 GB, from [ollama.com](https://ollama.com/library/gemma3n/tags)) ```sh ollama run gemma3n:e2b ``` Other versions: Q8_0 (~7.03 GB) `ollama run gemma3n:e2b-it-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 Gemma 3n E2B IT (4bit, needs ~4.5 GB, Mac only, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/gemma-3n-E2B-it-MLX-4bit/tree/ca848d026e284ba13bd93a8fa766833e2e97a644)) ```sh mlx_lm.chat --model lmstudio-community/gemma-3n-E2B-it-MLX-4bit ``` 3. Or start an OpenAI-compatible server at http://localhost:8080/v1 ```sh mlx_lm.server --model lmstudio-community/gemma-3n-E2B-it-MLX-4bit ``` Other versions: 8bit (~6.7 GB, Mac only) `mlx_lm.chat --model lmstudio-community/gemma-3n-E2B-it-MLX-8bit` Model repository and documentation: [huggingface.co/google](https://huggingface.co/google/gemma-3n-E2B-it) [interactive version](https://whatmodel.sh/model/gemma-3n-e2b) · [JSON](https://whatmodel.sh/api?format=json&all=1)