Gemma 4 E4B hardware requirements
What is Gemma 4 E4B?
Gemma 4 E4B is one of the two edge models in Google DeepMind's Gemma 4 family, released in April 2026 for laptops and mobile devices. The E stands for effective: the model card counts 4.5B effective parameters, and per-layer embedding tables used only for lookups bring the total to about 8B. It takes text, images and audio with up to a 128k context, keeps thinking off unless the system prompt starts with <|think|>, and is licensed under Apache 2.0.
About the model
- Parameters
- 8B
- Attention
- Sliding-window attention
- Max context
- 131,072 tokens
- Released
- 2026-03-02
- Model card
- Gemma 4 E4B on Hugging Face
Other sizes in this series
| Model | Parameters | Baseline quant | Memory needed (8k) | GeForce RTX 4060 | GeForce RTX 5060 Ti 16GB | GeForce RTX 3090 |
|---|---|---|---|---|---|---|
| Gemma 4 E2B | 5.12B | Q4_K_M | 2.1 GB | Runs great | Runs great | Runs great |
| Gemma 4 E4Bthis model | 8B | Q4_K_M | 3.7 GB | Runs great | Runs great | Runs great |
| Gemma 4 12B | 11.96B | Q4_K_M | 7.8 GB | Runs well | Runs great | Runs great |
| Gemma 4 26B-A4B | 26.54B total · 4B active (MoE) | Q4_K_M | 17.7 GB | Runs well | Runs well | Runs great |
| Gemma 4 31B | 31.27B | Q4_K_M | 19.5 GB | Runs slowly | Runs slowly | Runs great |
GGUF files
| Quant | File size | Quality | Published by |
|---|---|---|---|
| UD-Q2_K_XL | 3.76 GB | Heavy loss | Community GGUF · unsloth |
| IQ4_XS | 4.72 GB | Good balance | Community GGUF · unsloth |
| Q4_K_Mbaseline | 4.98 GB | Good balance | Community GGUF · unsloth |
| Q8_0 | 8.19 GB | Near-lossless | Community GGUF · unsloth |
File sizes do not include the vision projector (mmproj) that image input needs.
Some repositories publish only Unsloth Dynamic (UD) variants; sizes are those of the file you would download.
Memory needed by quant and context
| Quant | 4k | 8k | 16k | 32k | 64k | 128k |
|---|---|---|---|---|---|---|
| UD-Q2_K_XL | 2.8 | 2.9 | 3.1 | 3.5 | 4.3 | 6.0 |
| IQ4_XS | 3.4 | 3.5 | 3.7 | 4.1 | 4.9 | 6.6 |
| Q4_K_M | 3.6 | 3.7 | 3.9 | 4.4 | 5.2 | 6.9 |
| Q8_0 | 5.8 | 5.9 | 6.1 | 6.5 | 7.4 | 9.0 |
Which GPUs can run Gemma 4 E4B?
How to read the verdicts
- Runs great
- Fully on the GPU at 20 tok/s or more
- Runs well
- Fully on the GPU at 8–20 tok/s; MoE experts in system RAM at 20 tok/s or more; or at least 90% on the GPU at 8 tok/s or more
- Runs slowly
- 2–8 tok/s; CPU-only; less than 90% on the GPU; or MoE experts in system RAM below 20 tok/s
- Won't run
- Does not fit, or under 2 tok/s
Filter this list by budget, type and context in the calculator
Scroll sideways to see every column.
| Hardware | Verdict | Speed | Quant | Memory | Runs as | Context | Notes | Price |
|---|---|---|---|---|---|---|---|---|
| GeForce RTX 5090 | Runs great | 234.0 tok/s8k estimate 257.7 tok/s (180.4–335.0, theoretical estimate ±30%)measured (1 run, 512 context) | Q4_K_M | 4.3 / 32.0 GB | Full GPU | up to 128k | — | $4,200 street (as of 2026-08-09) |
| GeForce RTX 3090 Ti | Runs great | est. 171.0 tok/s119.7–222.3theoretical estimate ±30% | Q4_K_M | 4.3 / 24.0 GB | Full GPU | up to 128k | — | $1,999 launch MSRP |
| GeForce RTX 4090 | Runs great | est. 171.0 tok/s119.7–222.3theoretical estimate ±30% | Q4_K_M | 4.3 / 24.0 GB | Full GPU | up to 128k | — | $2,755 street (as of 2026-08-09) |
| GeForce RTX 5080 | Runs great | est. 164.6 tok/s115.3–214.0theoretical estimate ±30% | Q4_K_M | 4.3 / 16.0 GB | Full GPU | up to 128k | — | $1,256 street (as of 2026-08-09) |
| Radeon RX 7900 XTX | Runs great | est. 163.9 tok/s114.7–213.1theoretical estimate ±30% | Q4_K_M | 4.3 / 24.0 GB | Full GPU | up to 128k | — | $999 launch MSRP |
| GeForce RTX 3090 | Runs great | est. 161.4 tok/s113.0–209.9theoretical estimate ±30% | Q4_K_M | 4.3 / 24.0 GB | Full GPU | up to 128k | — | $1,050 street (as of 2026-08-09) |
| Apple M5 Ultra · 96 GB | Runs great | est. 159.1 tok/s111.4–206.8theoretical estimate ±30% | Q4_K_M | 5.7 / 67.2 GB | Unified memory | up to 128k | — | — |
| GeForce RTX 3080 12GB | Runs great | est. 158.2 tok/s110.7–205.6theoretical estimate ±30% | Q4_K_M | 4.3 / 12.0 GB | Full GPU | up to 128k | — | $799 launch MSRP |
| GeForce RTX 5070 Ti | Runs great | est. 156.0 tok/s109.2–202.8theoretical estimate ±30% | Q4_K_M | 4.3 / 16.0 GB | Full GPU | up to 128k | — | $949 street (as of 2026-08-09) |
| GeForce RTX 5080 Laptop | Runs great | est. 156.0 tok/s109.2–202.8theoretical estimate ±30% | Q4_K_M | 4.3 / 16.0 GB | Full GPU | up to 128k | — | — |
| GeForce RTX 5090 Laptop | Runs great | est. 156.0 tok/s109.2–202.8theoretical estimate ±30% | Q4_K_M | 4.3 / 24.0 GB | Full GPU | up to 128k | — | — |
| Radeon RX 7900 XT | Runs great | est. 143.5 tok/s100.4–186.5theoretical estimate ±30% | Q4_K_M | 4.3 / 20.0 GB | Full GPU | up to 128k | — | $899 launch MSRP |
| GeForce RTX 3080 10GB | Runs great | est. 136.7 tok/s95.7–177.7theoretical estimate ±30% | Q4_K_M | 4.3 / 10.0 GB | Full GPU | up to 128k | — | $699 launch MSRP |
| GeForce RTX 4080 Super | Runs great | est. 133.1 tok/s93.2–173.1theoretical estimate ±30% | Q4_K_M | 4.3 / 16.0 GB | Full GPU | up to 128k | — | $999 launch MSRP |
| GeForce RTX 4080 | Runs great | est. 130.3 tok/s91.2–169.4theoretical estimate ±30% | Q4_K_M | 4.3 / 16.0 GB | Full GPU | up to 128k | — | $1,199 launch MSRP |
| GeForce RTX 4070 Ti Super | Runs great | est. 123.5 tok/s86.4–160.5theoretical estimate ±30% | Q4_K_M | 4.3 / 16.0 GB | Full GPU | up to 128k | — | $799 launch MSRP |
| GeForce RTX 5070 | Runs great | est. 123.5 tok/s86.4–160.5theoretical estimate ±30% | Q4_K_M | 4.3 / 12.0 GB | Full GPU | up to 128k | — | $629 street (as of 2026-08-09) |
| GeForce RTX 5070 Ti Laptop | Runs great | est. 123.5 tok/s86.4–160.5theoretical estimate ±30% | Q4_K_M | 4.3 / 12.0 GB | Full GPU | up to 128k | — | — |
| Radeon RX 9070 | Runs great | est. 121.6 tok/s85.1–158.0theoretical estimate ±30% | Q4_K_M | 4.3 / 16.0 GB | Full GPU | up to 128k | — | $639 street (as of 2026-08-09) |
| Radeon RX 9070 XT | Runs great | est. 121.6 tok/s85.1–158.0theoretical estimate ±30% | Q4_K_M | 4.3 / 16.0 GB | Full GPU | up to 128k | — | $689 street (as of 2026-08-09) |
| Radeon RX 7800 XT | Runs great | est. 118.4 tok/s82.9–154.0theoretical estimate ±30% | Q4_K_M | 4.3 / 16.0 GB | Full GPU | up to 128k | — | $499 launch MSRP |
| Apple M3 Ultra · 96 GB | Runs great | est. 117.0 tok/s81.9–152.1theoretical estimate ±30% | Q4_K_M | 5.7 / 67.2 GB | Unified memory | up to 128k | — | — |
| Apple M5 Max (40-core GPU) · 48 GB | Runs great | est. 116.9 tok/s81.8–152.0theoretical estimate ±30% | Q4_K_M | 5.7 / 33.6 GB | Unified memory | up to 128k | — | — |
| Apple M2 Ultra · 64 GB | Runs great | est. 114.8 tok/s80.4–149.2theoretical estimate ±30% | Q4_K_M | 5.7 / 44.8 GB | Unified memory | up to 128k | — | — |
| GeForce RTX 2080 Ti | Runs great | est. 114.8 tok/s80.4–149.2theoretical estimate ±30% | Q4_K_M | 4.3 / 11.0 GB | Full GPU | up to 128k | — | $999 launch MSRP |
| GeForce RTX 4090 Laptop | Runs great | est. 108.4 tok/s75.9–141.0theoretical estimate ±30% | Q4_K_M | 4.3 / 16.0 GB | Full GPU | up to 128k | — | — |
| Apple M4 Max (40-core GPU) · 48 GB | Runs great | est. 106.4 tok/s74.5–138.3theoretical estimate ±30% | Q4_K_M | 5.7 / 33.6 GB | Unified memory | up to 128k | — | — |
| GeForce RTX 4070 | Runs great | est. 96.6 tok/s67.7–125.6theoretical estimate ±30% | Q4_K_M | 4.3 / 12.0 GB | Full GPU | up to 128k | — | $599 launch MSRP |
| GeForce RTX 4070 Super | Runs great | est. 96.6 tok/s67.7–125.6theoretical estimate ±30% | Q4_K_M | 4.3 / 12.0 GB | Full GPU | up to 128k | — | $599 launch MSRP |
| GeForce GTX 1080 Ti | Runs great | est. 93.3 tok/s65.3–121.3theoretical estimate ±30% | Q4_K_M | 4.3 / 11.0 GB | Full GPU | up to 128k | — | $699 launch MSRP |
| Apple M5 Max (32-core GPU) · 36 GB | Runs great | est. 92.3 tok/s64.6–120.0theoretical estimate ±30% | Q4_K_M | 5.7 / 25.2 GB | Unified memory | up to 128k | — | — |
| GeForce RTX 3070 | Runs great | est. 87.2 tok/s61.0–113.3theoretical estimate ±30% | Q4_K_M | 4.3 / 8.0 GB | Full GPU | up to 128k | — | $499 launch MSRP |
| GeForce RTX 5060 | Runs great | est. 87.2 tok/s61.0–113.3theoretical estimate ±30% | Q4_K_M | 4.3 / 8.0 GB | Full GPU | up to 128k | — | $339 street (as of 2026-08-09) |
| GeForce RTX 5060 Ti 16GB | Runs great | est. 87.2 tok/s61.0–113.3theoretical estimate ±30% | Q4_K_M | 4.3 / 16.0 GB | Full GPU | up to 128k | — | $569 street (as of 2026-08-09) |
| GeForce RTX 5060 Ti 8GB | Runs great | est. 87.2 tok/s61.0–113.3theoretical estimate ±30% | Q4_K_M | 4.3 / 8.0 GB | Full GPU | up to 128k | — | $429 street (as of 2026-08-09) |
| GeForce RTX 4080 Laptop | Runs great | est. 84.4 tok/s59.1–109.7theoretical estimate ±30% | Q4_K_M | 4.3 / 12.0 GB | Full GPU | up to 128k | — | — |
| Apple M4 Max (32-core GPU) · 36 GB | Runs great | est. 83.8 tok/s58.6–108.9theoretical estimate ±30% | Q4_K_M | 5.7 / 25.2 GB | Unified memory | up to 128k | — | — |
| Intel Arc B580 | Runs great | est. 73.8 tok/s51.7–96.0theoretical estimate ±30% | Q4_K_M | 4.3 / 12.0 GB | Full GPU | up to 128k | — | $290 street (as of 2026-08-09) |
| GeForce RTX 3060 12GB | Runs great | est. 71.7 tok/s50.2–93.2theoretical estimate ±30% | Q4_K_M | 4.3 / 12.0 GB | Full GPU | up to 128k | — | $250 street (as of 2026-08-09) |
| Radeon RX 9060 XT 16GB | Runs great | est. 67.9 tok/s47.6–88.3theoretical estimate ±30% | Q4_K_M | 4.3 / 16.0 GB | Full GPU | up to 128k | — | $449 street (as of 2026-08-09) |
| Apple M5 Pro · 24 GB | Runs great | est. 65.1 tok/s45.6–84.7theoretical estimate ±30% | Q4_K_M | 5.7 / 16.8 GB | Unified memory | up to 128k | — | — |
| NVIDIA DGX Spark · 64 GB | Runs great | est. 61.5 tok/s43.1–80.0theoretical estimate ±30% | Q4_K_M | 5.7 / 48.2 GB | Unified memory | up to 128k | — | $3,999 launch MSRP (128 GB) |
| Ryzen AI Max+ 395 (Strix Halo) · 32 GB | Runs great | est. 59.3 tok/s41.5–77.1theoretical estimate ±30% | Q4_K_M | 5.7 / 22.4 GB | Unified memory | up to 128k | — | $1,999 launch MSRP (128 GB) |
| Apple M4 Pro · 24 GB | Runs great | est. 58.7 tok/s41.1–76.3theoretical estimate ±30% | Q4_K_M | 5.7 / 16.8 GB | Unified memory | up to 128k | — | — |
| GeForce RTX 4060 Ti 16GB | Runs great | est. 58.5 tok/s40.9–76.0theoretical estimate ±30% | Q4_K_M | 4.3 / 16.0 GB | Full GPU | up to 128k | — | $499 launch MSRP |
| GeForce RTX 4060 Ti 8GB | Runs great | est. 58.5 tok/s40.9–76.0theoretical estimate ±30% | Q4_K_M | 4.3 / 8.0 GB | Full GPU | up to 128k | — | $399 launch MSRP |
| GeForce RTX 4060 | Runs great | est. 55.5 tok/s38.8–72.1theoretical estimate ±30% | Q4_K_M | 4.3 / 8.0 GB | Full GPU | up to 128k | — | $299 launch MSRP |
| GeForce RTX 4060 Laptop | Runs great | est. 52.5 tok/s36.7–68.2theoretical estimate ±30% | Q4_K_M | 4.3 / 8.0 GB | Full GPU | up to 128k | — | — |
| GeForce RTX 4070 Laptop | Runs great | est. 52.5 tok/s36.7–68.2theoretical estimate ±30% | Q4_K_M | 4.3 / 8.0 GB | Full GPU | up to 128k | — | — |
| GeForce RTX 3050 8GB | Runs great | est. 46.3 tok/s32.4–60.2theoretical estimate ±30% | Q4_K_M | 4.3 / 8.0 GB | Full GPU | up to 64k | — | $249 launch MSRP |
| Apple M5 · 16 GB | Runs great | est. 34.6 tok/s24.2–44.9theoretical estimate ±30% | Q4_K_M | 5.7 / 11.2 GB | Unified memory | up to 32k | — | — |
| Apple M6 · 16 GB | Runs great | est. 34.6 tok/s24.2–44.9theoretical estimate ±30% | Q4_K_M | 5.7 / 11.2 GB | Unified memory | up to 32k | — | — |
| Apple M4 · 16 GB | Runs well | est. 27.4 tok/s19.1–35.6theoretical estimate ±30% | Q4_K_M | 5.7 / 11.2 GB | Unified memory | up to 128k |
| — |
| DDR5-6000 dual-channel CPU · 16 GB | Runs slowly | est. 15.1 tok/s10.6–19.7theoretical estimate ±30% | Q4_K_M | 5.1 GB RAM | CPU only | up to 128k |
| — |
| DDR5-5600 dual-channel CPU · 16 GB | Runs slowly | est. 14.1 tok/s9.9–18.3theoretical estimate ±30% | Q4_K_M | 5.1 GB RAM | CPU only | up to 128k |
| — |
| DDR4-3200 dual-channel CPU · 16 GB | Runs slowly | est. 8.1 tok/s5.6–10.5theoretical estimate ±30% | Q4_K_M | 5.1 GB RAM | CPU only | up to 128k | — | — |
Models with a thinking mode spend extra tokens thinking, so their speed thresholds are 1.5× stricter (30 / 12 / 3 tok/s). With thinking turned off, a model may do one grade better.
Cheapest GPUs that run it
- Runs greatGeForce RTX 3060 12GB$250 street (as of 2026-08-09), est. 71.7 tok/s (50.2–93.2, theoretical estimate ±30%).
Measured results for Gemma 4 E4B
Public benchmarks we calibrate against. Their conditions (context, backend, flags) can differ from the estimates above.
| Hardware | Quant | Backend | Context | Prompt (tok/s) | Generation (tok/s) | Flags | Source | Measured |
|---|---|---|---|---|---|---|---|---|
| GeForce RTX 5090 | Q4_K_M | llama.cpp | 512 | — | 234.0 | Vulkan | github.com | 2026-09-08 |
| NVIDIA GeForce RTX 4070 Ti 12GB (Ryzen 9 7950X3D) | Q8_0 | llama.cpp | 512 | 6,757.0 | 69.7 | — | alfonsofortunato.com | 2026-04-04 |
Frequently asked questions
How much VRAM does Gemma 4 E4B need?
At Q4_K_M the file is 4.98 GB; with 8k context a graphics card needs about 3.7 GB (weights 3.03 GB, KV cache 0.1 GB, compute buffer 0.6 GB). The smallest tracked file (UD-Q2_K_XL (heavy quality loss)) needs about 2.9 GB. Part of the file stays off the graphics card: llama.cpp keeps the input embedding table in system RAM. A Mac counts the whole file, so it needs a little more.
What is the cheapest GPU that runs Gemma 4 E4B well?
Cheapest that runs it great: GeForce RTX 3060 12GB at $250 (street price as of 2026-08-09), est. 71.7 tok/s (50.2–93.2, theoretical estimate ±30%).
Can Gemma 4 E4B run on an 8 GB, 16 GB or 24 GB GPU?
At Q4_K_M with 8k context: GeForce RTX 4060 — Runs great, est. 55.5 tok/s (38.8–72.1, theoretical estimate ±30%); GeForce RTX 5060 Ti 16GB — Runs great, est. 87.2 tok/s (61.0–113.3, theoretical estimate ±30%); GeForce RTX 3090 — Runs great, est. 161.4 tok/s (113.0–209.9, theoretical estimate ±30%).
Speeds are estimates from memory bandwidth, calibrated against public benchmarks, and each one comes with an error band and a confidence label. Real results vary with drivers, backend, context length and thermals.