whatmodel.sh / Gemma 3n E2B IT
Gemma 3n E2B IT
↗ share
+ contribute
Gemma 3n E2B IT is an open-weights language model from Google, released June 2025. Its published size is 5.44B parameters (2B active per token) and a 32k-token context window. Licence: gemma.
42.9±1.6#482 of 719 LLMsfrom 12 benchmark results across 4 sources
$ how it compares
The LLMs ranked either side of Gemma 3n E2B IT, out of 828 with a score, then more from Google.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 569 |  CogVLM2 Llama3 Chat 19B | 43±3 | local | ||
| 570 |  Code Llama 70B (base) | 43±1 | local | ||
| 571 | Gemma 3n E2B IT | 43±2 | local | ||
| 572 |  Reka Flash 21B (2024-02-26) | 43±1 | comparison only | ||
| 573 |  T5 Small | 43±9 | local | ||
| more from Google | |||||
| 9 |  Gemini 4 Argon (limited preview) | 82±1 | limited preview | ||
| 23 | Gemini 3.8 Flash | 77±1 | api $1.50/M | ||
| 32 | Gemini 3.7 Flash | 76±1 | api $1.50/M | ||
$ what it's good at
coding█··········█ #2 of 2 · 29% of its score · 3 tests
math█··········█ #2 of 2 * · 24% of its score · 2 tests
other tests9 not in a task · 45% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| MMLU lab | 60.1% ✎ | ▁▁▁▁▁▂▃▅█▃▆▃ beats 70/183 | 10% |
| Open Persian LLM Leaderboard v2 average (21 tasks) | 52.0% ✎ | ▂▄▂▃█▄▃▆▃▅▂▂ beats 36/81 | 10% |
| MMLU-Pro lab | 40.5% ✎ | ▁▁·▁▁▁▃▂▄▅█▁ beats 27/188 | 8.6% |
| Global-MMLU lab | 55.1% ✎ | ▃▅▃·▃···▅·▅█ beats 0/11 | 5.1% |
| Open PT-LLM Leaderboard average (9 Portuguese tasks) | 56.3% ✎ | ▃▁▁▁▁▁▁▁▃█▅▁ beats 80/276 | 4.9% |
| EuroEval HellaSwag (English), test split, 5-shot generative MC, accuracy | 66.4% ✎ | ▃·▁▁▂▁▂▅█▇▄▁ beats 42/100 | 3.7% |
| EuroEval Life in the UK (English knowledge MC), test split, 5-shot, accuracy | 79.4% ✎ | ▁▁▁▁▁▁▁▂▄█▄▃ beats 63/146 | 3.2% |
| GPQA Diamond (Google RelaxedAccuracy/accuracy) lab | 24.8% ✎ | ······█····· beats 1/1 | — |
1 more test
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Codegolf v2.2 lab | 11.0% ✎ | █··········█ beats 0/1 | — |
$ where to run it
llama.cpp · recommended
version
- Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.macOS / Linux
curl -LsSf https://llama.app/install.sh | shWindowsirm https://llama.app/install.ps1 | iex - Run Gemma 3n E2B IT · needs ~3 GB · from huggingface.co/unslothchat in the terminal
llama cli -hf unsloth/gemma-3n-E2B-it-GGUF:Q4_K_M - Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf unsloth/gemma-3n-E2B-it-GGUF:Q4_K_M
LM Studio
version
- Install LM Studio · Open the app once before using the lms command.
- Run Gemma 3n E2B IT · needs ~3 GB · from huggingface.co/unsloth
- Or download it in the terminal
lms get https://huggingface.co/unsloth/gemma-3n-E2B-it-GGUF@Q4_K_M -y - Then run lms ls to see its name, and lms chat followed by that name
Ollama
version
- Install Ollama
download it (macOS / Windows) ↗
Linuxcurl -fsSL https://ollama.com/install.sh | sh - Run Gemma 3n E2B IT · needs ~3.76 GB · from ollama.comchat in the terminal
ollama run gemma3n:e2b
MLX (Mac)
version
- Install MLX (Mac) · Apple silicon Macs only, text only.with uv
uv tool install mlx-lmorpip install mlx-lm - Run Gemma 3n E2B IT · needs ~4.5 GB, Mac only · from huggingface.co/lmstudio-communitychat in the terminal
mlx_lm.chat --model lmstudio-community/gemma-3n-E2B-it-MLX-4bit - Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model lmstudio-community/gemma-3n-E2B-it-MLX-4bit
Official weights and documentation: huggingface.co/google ↗
$ how the score was worked out · Google DeepMind Gemma 3n E2B… 78%, Open Persian LLM Leaderboard… 10%, EuroEval leaderboards (result… 7% and 1 more
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The 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's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 Google DeepMind Gemma 3n E2B IT model card · lab-reported | 8 | 78% |
| Open Persian LLM Leaderboard v2 (Part DP AI and AUT NLP Lab) | 1 | 10% |
 EuroEval leaderboards (results collection), English generative leaderboard | 2 | 7% |
| Open Portuguese LLM Leaderboard (archived) | 1 | 5% |
| prior · pull toward the middle | 1% | |
| weighted average | = 42.9 |
not counted (2)
| Google DeepMind Gemma 3n E2B IT model card | GPQA Diamond (Google RelaxedAccuracy/accuracy) [correct] | 24.8% | fewer than three models have this test yet, or this source ran it for this model alone |
| Google DeepMind Gemma 3n E2B IT model card | Codegolf v2.2 [correct] | 11.0% | fewer than three models have this test yet, or this source ran it for this model alone |
$ improve this page
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json