Gemma 4 E2B hardware requirements
What is Gemma 4 E2B?
Gemma 4 E2B is the smaller of the two edge models in Google DeepMind's Gemma 4 family, released in April 2026 for laptops and mobile devices. Its model card counts 2.3B effective parameters, and per-layer embedding tables used only for lookups bring the total to about 5.1B. Like E4B, 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
- 5.12B
- Attention
- Sliding-window attention
- Max context
- 131,072 tokens
- Released
- 2026-03-02
- Model card
- Gemma 4 E2B 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 E2Bthis model | 5.12B | Q4_K_M | 2.1 GB | Runs great | Runs great | Runs great |
| Gemma 4 E4B | 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 | 2.40 GB | Heavy loss | Community GGUF · unsloth |
| IQ4_XS | 2.98 GB | Good balance | Community GGUF · unsloth |
| Q4_K_Mbaseline | 3.11 GB | Good balance | Community GGUF · unsloth |
| Q8_0 | 5.05 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 | 1.6 | 1.7 | 1.8 | 2.1 | 2.6 | 3.6 |
| IQ4_XS | 1.9 | 2.0 | 2.1 | 2.4 | 2.9 | 3.9 |
| Q4_K_M | 2.0 | 2.1 | 2.2 | 2.5 | 3.0 | 4.0 |
| Q8_0 | 3.1 | 3.2 | 3.3 | 3.5 | 4.0 | 5.0 |
Which GPUs can run Gemma 4 E2B?
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 | 341.8 tok/s8k estimate 386.7 tok/s (270.7–502.7, theoretical estimate ±30%)measured (1 run, 512 context) | Q4_K_M | 2.7 / 32.0 GB | Full GPU | up to 128k | — | $4,200 street (as of 2026-08-09) |
| GeForce RTX 3090 Ti | Runs great | est. 281.9 tok/s197.3–366.5theoretical estimate ±30% | Q4_K_M | 2.7 / 24.0 GB | Full GPU | up to 128k | — | $1,999 launch MSRP |
| GeForce RTX 4090 | Runs great | est. 281.9 tok/s197.3–366.5theoretical estimate ±30% | Q4_K_M | 2.7 / 24.0 GB | Full GPU | up to 128k | — | $2,755 street (as of 2026-08-09) |
| GeForce RTX 5080 | Runs great | est. 273.4 tok/s191.4–355.5theoretical estimate ±30% | Q4_K_M | 2.7 / 16.0 GB | Full GPU | up to 128k | — | $1,256 street (as of 2026-08-09) |
| GeForce RTX 3090 | Runs great | est. 269.1 tok/s188.4–349.8theoretical estimate ±30% | Q4_K_M | 2.7 / 24.0 GB | Full GPU | up to 128k | — | $1,050 street (as of 2026-08-09) |
| GeForce RTX 3080 12GB | Runs great | est. 264.7 tok/s185.3–344.1theoretical estimate ±30% | Q4_K_M | 2.7 / 12.0 GB | Full GPU | up to 128k | — | $799 launch MSRP |
| GeForce RTX 5070 Ti | Runs great | est. 261.7 tok/s183.2–340.2theoretical estimate ±30% | Q4_K_M | 2.7 / 16.0 GB | Full GPU | up to 128k | — | $949 street (as of 2026-08-09) |
| GeForce RTX 5080 Laptop | Runs great | est. 261.7 tok/s183.2–340.2theoretical estimate ±30% | Q4_K_M | 2.7 / 16.0 GB | Full GPU | up to 128k | — | — |
| GeForce RTX 5090 Laptop | Runs great | est. 261.7 tok/s183.2–340.2theoretical estimate ±30% | Q4_K_M | 2.7 / 24.0 GB | Full GPU | up to 128k | — | — |
| Radeon RX 7900 XTX | Runs great | est. 258.1 tok/s180.6–335.5theoretical estimate ±30% | Q4_K_M | 2.7 / 24.0 GB | Full GPU | up to 128k | — | $999 launch MSRP |
| Apple M5 Ultra · 96 GB | Runs great | est. 252.2 tok/s176.5–327.9theoretical estimate ±30% | Q4_K_M | 3.7 / 67.2 GB | Unified memory | up to 128k | — | — |
| GeForce RTX 3080 10GB | Runs great | est. 234.5 tok/s164.2–304.9theoretical estimate ±30% | Q4_K_M | 2.7 / 10.0 GB | Full GPU | up to 128k | — | $699 launch MSRP |
| Radeon RX 7900 XT | Runs great | est. 232.6 tok/s162.8–302.3theoretical estimate ±30% | Q4_K_M | 2.7 / 20.0 GB | Full GPU | up to 128k | — | $899 launch MSRP |
| GeForce RTX 4080 Super | Runs great | est. 229.4 tok/s160.6–298.2theoretical estimate ±30% | Q4_K_M | 2.7 / 16.0 GB | Full GPU | up to 128k | — | $999 launch MSRP |
| GeForce RTX 4080 | Runs great | est. 225.3 tok/s157.7–292.9theoretical estimate ±30% | Q4_K_M | 2.7 / 16.0 GB | Full GPU | up to 128k | — | $1,199 launch MSRP |
| GeForce RTX 4070 Ti Super | Runs great | est. 215.2 tok/s150.7–279.8theoretical estimate ±30% | Q4_K_M | 2.7 / 16.0 GB | Full GPU | up to 128k | — | $799 launch MSRP |
| GeForce RTX 5070 | Runs great | est. 215.2 tok/s150.7–279.8theoretical estimate ±30% | Q4_K_M | 2.7 / 12.0 GB | Full GPU | up to 128k | — | $629 street (as of 2026-08-09) |
| GeForce RTX 5070 Ti Laptop | Runs great | est. 215.2 tok/s150.7–279.8theoretical estimate ±30% | Q4_K_M | 2.7 / 12.0 GB | Full GPU | up to 128k | — | — |
| Radeon RX 9070 | Runs great | est. 203.6 tok/s142.5–264.6theoretical estimate ±30% | Q4_K_M | 2.7 / 16.0 GB | Full GPU | up to 128k | — | $639 street (as of 2026-08-09) |
| Radeon RX 9070 XT | Runs great | est. 203.6 tok/s142.5–264.6theoretical estimate ±30% | Q4_K_M | 2.7 / 16.0 GB | Full GPU | up to 128k | — | $689 street (as of 2026-08-09) |
| GeForce RTX 2080 Ti | Runs great | est. 202.2 tok/s141.5–262.9theoretical estimate ±30% | Q4_K_M | 2.7 / 11.0 GB | Full GPU | up to 128k | — | $999 launch MSRP |
| Radeon RX 7800 XT | Runs great | est. 199.2 tok/s139.5–259.0theoretical estimate ±30% | Q4_K_M | 2.7 / 16.0 GB | Full GPU | up to 128k | — | $499 launch MSRP |
| Apple M3 Ultra · 96 GB | Runs great | est. 197.2 tok/s138.0–256.4theoretical estimate ±30% | Q4_K_M | 3.7 / 67.2 GB | Unified memory | up to 128k | — | — |
| Apple M5 Max (40-core GPU) · 48 GB | Runs great | est. 197.1 tok/s138.0–256.3theoretical estimate ±30% | Q4_K_M | 3.7 / 33.6 GB | Unified memory | up to 128k | — | — |
| Apple M2 Ultra · 64 GB | Runs great | est. 194.2 tok/s135.9–252.4theoretical estimate ±30% | Q4_K_M | 3.7 / 44.8 GB | Unified memory | up to 128k | — | — |
| GeForce RTX 4090 Laptop | Runs great | est. 192.5 tok/s134.7–250.2theoretical estimate ±30% | Q4_K_M | 2.7 / 16.0 GB | Full GPU | up to 128k | — | — |
| Apple M4 Max (40-core GPU) · 48 GB | Runs great | est. 182.3 tok/s127.6–237.0theoretical estimate ±30% | Q4_K_M | 3.7 / 33.6 GB | Unified memory | up to 128k | — | — |
| GeForce RTX 4070 | Runs great | est. 174.1 tok/s121.9–226.3theoretical estimate ±30% | Q4_K_M | 2.7 / 12.0 GB | Full GPU | up to 128k | — | $599 launch MSRP |
| GeForce RTX 4070 Super | Runs great | est. 174.1 tok/s121.9–226.3theoretical estimate ±30% | Q4_K_M | 2.7 / 12.0 GB | Full GPU | up to 128k | — | $599 launch MSRP |
| GeForce GTX 1080 Ti | Runs great | est. 168.7 tok/s118.1–219.4theoretical estimate ±30% | Q4_K_M | 2.7 / 11.0 GB | Full GPU | up to 128k | — | $699 launch MSRP |
| Apple M5 Max (32-core GPU) · 36 GB | Runs great | est. 161.7 tok/s113.2–210.2theoretical estimate ±30% | Q4_K_M | 3.7 / 25.2 GB | Unified memory | up to 128k | — | — |
| GeForce RTX 3070 | Runs great | est. 158.9 tok/s111.2–206.6theoretical estimate ±30% | Q4_K_M | 2.7 / 8.0 GB | Full GPU | up to 128k | — | $499 launch MSRP |
| GeForce RTX 5060 | Runs great | est. 158.9 tok/s111.2–206.6theoretical estimate ±30% | Q4_K_M | 2.7 / 8.0 GB | Full GPU | up to 128k | — | $339 street (as of 2026-08-09) |
| GeForce RTX 5060 Ti 16GB | Runs great | est. 158.9 tok/s111.2–206.6theoretical estimate ±30% | Q4_K_M | 2.7 / 16.0 GB | Full GPU | up to 128k | — | $569 street (as of 2026-08-09) |
| GeForce RTX 5060 Ti 8GB | Runs great | est. 158.9 tok/s111.2–206.6theoretical estimate ±30% | Q4_K_M | 2.7 / 8.0 GB | Full GPU | up to 128k | — | $429 street (as of 2026-08-09) |
| GeForce RTX 4080 Laptop | Runs great | est. 154.4 tok/s108.1–200.7theoretical estimate ±30% | Q4_K_M | 2.7 / 12.0 GB | Full GPU | up to 128k | — | — |
| Apple M4 Max (32-core GPU) · 36 GB | Runs great | est. 148.7 tok/s104.1–193.3theoretical estimate ±30% | Q4_K_M | 3.7 / 25.2 GB | Unified memory | up to 128k | — | — |
| GeForce RTX 3060 12GB | Runs great | est. 133.3 tok/s93.3–173.3theoretical estimate ±30% | Q4_K_M | 2.7 / 12.0 GB | Full GPU | up to 128k | — | $250 street (as of 2026-08-09) |
| Intel Arc B580 | Runs great | est. 133.1 tok/s93.2–173.0theoretical estimate ±30% | Q4_K_M | 2.7 / 12.0 GB | Full GPU | up to 128k | — | $290 street (as of 2026-08-09) |
| Radeon RX 9060 XT 16GB | Runs great | est. 123.7 tok/s86.6–160.8theoretical estimate ±30% | Q4_K_M | 2.7 / 16.0 GB | Full GPU | up to 128k | — | $449 street (as of 2026-08-09) |
| Apple M5 Pro · 24 GB | Runs great | est. 119.1 tok/s83.4–154.9theoretical estimate ±30% | Q4_K_M | 3.7 / 16.8 GB | Unified memory | up to 128k | — | — |
| NVIDIA DGX Spark · 64 GB | Runs great | est. 115.9 tok/s81.1–150.6theoretical estimate ±30% | Q4_K_M | 3.7 / 48.2 GB | Unified memory | up to 128k | — | $3,999 launch MSRP (128 GB) |
| GeForce RTX 4060 Ti 16GB | Runs great | est. 110.6 tok/s77.4–143.8theoretical estimate ±30% | Q4_K_M | 2.7 / 16.0 GB | Full GPU | up to 128k | — | $499 launch MSRP |
| GeForce RTX 4060 Ti 8GB | Runs great | est. 110.6 tok/s77.4–143.8theoretical estimate ±30% | Q4_K_M | 2.7 / 8.0 GB | Full GPU | up to 128k | — | $399 launch MSRP |
| Ryzen AI Max+ 395 (Strix Halo) · 32 GB | Runs great | est. 109.5 tok/s76.6–142.3theoretical estimate ±30% | Q4_K_M | 3.7 / 22.4 GB | Unified memory | up to 128k | — | $1,999 launch MSRP (128 GB) |
| Apple M4 Pro · 24 GB | Runs great | est. 108.4 tok/s75.9–141.0theoretical estimate ±30% | Q4_K_M | 3.7 / 16.8 GB | Unified memory | up to 128k | — | — |
| GeForce RTX 4060 | Runs great | est. 105.3 tok/s73.7–137.0theoretical estimate ±30% | Q4_K_M | 2.7 / 8.0 GB | Full GPU | up to 128k | — | $299 launch MSRP |
| GeForce RTX 4060 Laptop | Runs great | est. 100.0 tok/s70.0–130.0theoretical estimate ±30% | Q4_K_M | 2.7 / 8.0 GB | Full GPU | up to 128k | — | — |
| GeForce RTX 4070 Laptop | Runs great | est. 100.0 tok/s70.0–130.0theoretical estimate ±30% | Q4_K_M | 2.7 / 8.0 GB | Full GPU | up to 128k | — | — |
| GeForce RTX 3050 8GB | Runs great | est. 89.0 tok/s62.3–115.7theoretical estimate ±30% | Q4_K_M | 2.7 / 8.0 GB | Full GPU | up to 128k | — | $249 launch MSRP |
| Apple M5 · 16 GB | Runs great | est. 66.5 tok/s46.6–86.5theoretical estimate ±30% | Q4_K_M | 3.7 / 11.2 GB | Unified memory | up to 128k | — | — |
| Apple M6 · 16 GB | Runs great | est. 66.5 tok/s46.6–86.5theoretical estimate ±30% | Q4_K_M | 3.7 / 11.2 GB | Unified memory | up to 128k | — | — |
| Apple M4 · 16 GB | Runs great | est. 53.3 tok/s37.3–69.3theoretical estimate ±30% | Q4_K_M | 3.7 / 11.2 GB | Unified memory | up to 128k | — | — |
| DDR5-6000 dual-channel CPU · 16 GB | Runs slowly | est. 31.0 tok/s21.7–40.2theoretical estimate ±30% | Q4_K_M | 3.2 GB RAM | CPU only | up to 128k |
| — |
| DDR5-5600 dual-channel CPU · 16 GB | Runs slowly | est. 28.9 tok/s20.2–37.6theoretical estimate ±30% | Q4_K_M | 3.2 GB RAM | CPU only | up to 128k |
| — |
| DDR4-3200 dual-channel CPU · 16 GB | Runs slowly | est. 16.5 tok/s11.6–21.5theoretical estimate ±30% | Q4_K_M | 3.2 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. 133.3 tok/s (93.3–173.3, theoretical estimate ±30%).
Measured results for Gemma 4 E2B
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 | — | 341.8 | Vulkan | github.com | 2026-09-08 |
Frequently asked questions
How much VRAM does Gemma 4 E2B need?
At Q4_K_M the file is 3.11 GB; with 8k context a graphics card needs about 2.1 GB (weights 1.48 GB, KV cache 0.1 GB, compute buffer 0.6 GB). The smallest tracked file (UD-Q2_K_XL (heavy quality loss)) needs about 1.7 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 E2B well?
Cheapest that runs it great: GeForce RTX 3060 12GB at $250 (street price as of 2026-08-09), est. 133.3 tok/s (93.3–173.3, theoretical estimate ±30%).
Can Gemma 4 E2B run on an 8 GB, 16 GB or 24 GB GPU?
At Q4_K_M with 8k context: GeForce RTX 4060 — Runs great, est. 105.3 tok/s (73.7–137.0, theoretical estimate ±30%); GeForce RTX 5060 Ti 16GB — Runs great, est. 158.9 tok/s (111.2–206.6, theoretical estimate ±30%); GeForce RTX 3090 — Runs great, est. 269.1 tok/s (188.4–349.8, 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.