gemma-3n-E4B-it
by unsloth · built on gemma 3n E4B · Official #21 of 32 by downloads
Benchmark scores
Official models are not scored on the UGI Leaderboard. We show downloads and likes instead. Estimated speed: ~12–22 tok/s (estimate), a rough range for a Q4_K_M file of this size on a recent flagship chip (Snapdragon 8 Gen 3 or A17 Pro class). It is not a measurement from this model.
About this model
gemma-3n-E4B-it is a 4B-parameter language model by unsloth, derived from google/gemma-3n-E4B using official GGUF conversion. You can download it and run it on your phone for free.
An official 4B release based on gemma 3n E4B, intended for everyday chat and writing on a phone. The Q4_K_M file is 4.54 GB.
Technically, it uses the gemma3n architecture, supports a context window of 32,768 tokens (a phone will use far less).
Which RAM sizes can run it?
| Device RAM | iOS-safe | Android-safe |
|---|---|---|
| 4 GB RAM | Too big | Too big |
| 6 GB RAM | Too big | Fits with Q3_K_M (~3.69 GB) |
| 8 GB RAM | Fits with Q4_K_M (~4.54 GB) | Fits with Q4_K_M (~4.54 GB) |
| 12 GB RAM | Fits with Q4_K_M (~4.54 GB) | Fits with Q4_K_M (~4.54 GB) |
| 16 GB+ RAM | Fits with Q4_K_M (~4.54 GB) | Fits with Q4_K_M (~4.54 GB) |
iOS-safe minimum: 8 GB. Android-safe minimum: 6 GB (using Q3_K_M). On iOS we treat about 57% of device RAM as usable (in the 55–60% range), because the system keeps a large share and may evict an app that goes higher. On Android we treat about 67% of device RAM as usable (in the 65–70% range), leaving room for the system and the app. See the guide.
Run it on your phone
Compatible apps:
iOS
- PocketPal AI — Free on the App Store. Can search Hugging Face and load a GGUF file on the device.
- Layla — On the App Store as a paid app. Load a custom GGUF from files on the device.
Android
- PocketPal AI — Free on Google Play. Can search Hugging Face and load a GGUF file on the device.
- Layla — On Google Play. Load a custom GGUF from storage. A limited free APK is also offered on the developer site.
- ChatterUI — Free and open source. Install the APK from the GitHub releases page. iOS is not available.
Hugging Face repository ID:
unsloth/gemma-3n-E4B-it-GGUFRecommended file: Q4_K_M — gemma-3n-E4B-it-Q4_K_M.gguf (about 4.54 GB)
- Install one of the apps above for your operating system. PocketPal AI is free on both iOS and Android and can search Hugging Face directly.
- In the app, add a model from Hugging Face and paste this repository ID:
unsloth/gemma-3n-E4B-it-GGUF - Download the Q4_K_M file (
gemma-3n-E4B-it-Q4_K_M.gguf) (about 4.54 GB), or Q3_K_M (about 3.69 GB) if memory is tight. Use wifi. This is the only step that needs internet. - Load the file and start chatting. Switch on airplane mode if you like: it keeps working. A first prompt to try:
I'm considering a decision this week. Give me the strongest case for it, the strongest case against it, and the one fact that would change my mind. Keep it short.
If the app is killed or feels slow, close other apps, try the smaller file, or pick a model with a lower iOS-safe or Android-safe minimum. On a computer, the same files work in LM Studio, Ollama and llama.cpp.
Available GGUF files (24)
| Quantization | Size | Download |
|---|---|---|
| IQ2_XXS gemma-3n-E4B-it-UD-IQ2_XXS.gguf | 2.83 GB | Download |
| IQ2_M gemma-3n-E4B-it-UD-IQ2_M.gguf | 3.11 GB | Download |
| Q2_K gemma-3n-E4B-it-Q2_K.gguf | 3.19 GB | Download |
| Q2_K_L gemma-3n-E4B-it-Q2_K_L.gguf | 3.19 GB | Download |
| IQ3_XXS gemma-3n-E4B-it-UD-IQ3_XXS.gguf | 3.35 GB | Download |
| Q3_K_S gemma-3n-E4B-it-Q3_K_S.gguf | 3.50 GB | Download |
| Q2_K gemma-3n-E4B-it-UD-Q2_K_XL.gguf | 3.66 GB | Download |
| Q3_K_MSMALLER gemma-3n-E4B-it-Q3_K_M.gguf | 3.69 GB | Download |
| Q3_K gemma-3n-E4B-it-UD-Q3_K_XL.gguf | 4.13 GB | Download |
| IQ4_XS gemma-3n-E4B-it-IQ4_XS.gguf | 4.27 GB | Download |
| IQ4_NL gemma-3n-E4B-it-IQ4_NL.gguf | 4.39 GB | Download |
| Q4_0 gemma-3n-E4B-it-Q4_0.gguf | 4.40 GB | Download |
| Q4_K_S gemma-3n-E4B-it-Q4_K_S.gguf | 4.40 GB | Download |
| Q4_K_MRECOMMENDED gemma-3n-E4B-it-Q4_K_M.gguf | 4.54 GB | Download |
| Q4_1 gemma-3n-E4B-it-Q4_1.gguf | 4.63 GB | Download |
| Q5_K_S gemma-3n-E4B-it-Q5_K_S.gguf | 4.95 GB | Download |
| Q5_K_M gemma-3n-E4B-it-Q5_K_M.gguf | 5.02 GB | Download |
| Q4_K gemma-3n-E4B-it-UD-Q4_K_XL.gguf | 5.39 GB | Download |
| Q5_K gemma-3n-E4B-it-UD-Q5_K_XL.gguf | 5.86 GB | Download |
| Q6_K gemma-3n-E4B-it-Q6_K.gguf | 6.27 GB | Download |
| Q6_K gemma-3n-E4B-it-UD-Q6_K_XL.gguf | 6.62 GB | Download |
| Q8_0 gemma-3n-E4B-it-Q8_0.gguf | 7.35 GB | Download |
| Q8_K gemma-3n-E4B-it-UD-Q8_K_XL.gguf | 10.59 GB | Download |
| F16 gemma-3n-E4B-it-F16.gguf | 13.74 GB | Download |
Direct downloads from unsloth/gemma-3n-E4B-it-GGUF on Hugging Face. Lower Q numbers are smaller and slightly less accurate.