Gemma 4 12B IT Assistant — Hardware Requirements & GPU Compatibility
Chat
Gemma 4 12B IT Assistant is a 12B-parameter open language model from Google in the Gemma 4 family. It supports a context window of up to 262,144 tokens. At Q4_K_M it needs about 7.52 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 32K)Best on 24 GB (Q4_K_M · ~87 tok/s · 256K)Macs from 16 GB (Q4_K_M · ~19 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Gemma 4 12B IT Assistant
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
vllm serve google/gemma-4-12B-it-assistant --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3080 Ti (12 GB) (4 GB headroom) · Q4_K_M
- Generation speed
- ~79 tok/s
- generation speed
- Cost per 1M output tokens
- $0.43
- per 1M output tokens
How Much VRAM Does Gemma 4 12B IT Assistant Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 5.4 GB | 5.7 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 256K context · ~121 tok/s | |
| 6.2 GB | 6.4 GB | 3-bit medium — 3-bit medium quantization | Fits: 256K context · ~106 tok/s | |
| 7.5 GB | 7.8 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 256K context · ~87 tok/s | |
| 8.9 GB | 9.1 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 256K context · ~74 tok/s | |
| 10.2 GB | 10.5 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 256K context · ~64 tok/s | |
| 12.3 GB | 12.6 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 256K context · ~53 tok/s | |
| 24.3 GB | 24.6 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Does not fit: needs 24.4 GB |
Selected Q4_K_M
4-bit medium quantization7.6GBat 16K context
- Weights
- 7.2 GB
- KV cache
- 0.1 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 16.4 GB free · ~87 tok/s
≥50 tok/s needs ≥578 GB/s · RTX 5090 ~155, RTX 3090 Ti ~87
Which Hardware Can Run Gemma 4 12B IT Assistant?
Q4_K_M · 7.5 GBGemma 4 12B IT Assistant (Q4_K_M) requires 7.5 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 10+ GB is recommended. Using the full 262K context window can add up to 2.1 GB, bringing total usage to 9.7 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3080. 151 devices with unified memory can run Gemma 4 12B IT Assistant, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), NVIDIA Jetson Orin Nano 8GB (Super).
≤ 11 GB
≤ 11 GB tier →fits · 0.4 GB free · up to 32K ctx
16 GPUs · 1 device family
RTX 3080~66Simulate output on RTX 3080RTX 2080 Ti~53Simulate output on RTX 2080 TiRTX 3070 Ti~53Simulate output on RTX 3070 TiOrin Nano~9Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)20 GB
fits · 12.4 GB free · up to 256K ctx
1 GPU
RX 7900 XT~64Simulate output on RX 7900 XT32–40 GB
fits · 24.4 GB free · up to 256K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~155Simulate output on RTX 5090M5 Max36 GB~43Simulate output on Mac Studio M5 Max (36 GB)M532 GB~14Simulate output on MacBook Air 13" M5 (32 GB)Orin~18Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~12Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to Gemma 4 12B IT Assistant
Gemma 4 family · 35 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Gemma 3 12B ITGoogle | 74.2 | 10.9 GB | ~82 | same size |
| Mistral Nemo Base 2407Mistral AI | — | 11 GB | ~81 | same size |
| Bielik 11B V3.0 Instructspeakleash | — | 27.4 GB | —needs > 24 GB | same size |
| NVIDIA Nemotron Nano 12B v2NVIDIA | — | 11 GB | ~81 | same size |
| Midm 2.0 Base InstructK-intelligence | — | 10.5 GB | ~86 | same size |
| MN 12B Mag Mell R1inflatebot | — | 11 GB | ~81 | same size |
Fine-tunes & variants(1)
Specifications
- Publisher
- Family
- Gemma 4
- Parameters
- 12B
- Architecture
- Gemma4UnifiedAssistantForCausalLM
- Context Length
- 262,144 tokens
- Vocabulary Size
- 262,144
- Release Date
- 2026-05-23
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Gemma 4 12B IT Assistant need?
Gemma 4 12B IT Assistant requires 7.5 GB of VRAM at Q4_K_M, or 24.3 GB at BF16. Full 262K context adds up to 2.1 GB (9.7 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 12B × 4.8 bits ÷ 8 = 7.2 GB
KV Cache + Overhead ≈ 0.3 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 2.5 GB (at full 262K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 4090 run Gemma 4 12B IT Assistant?
Yes, at Q8_0 (12.3 GB) or lower. Higher quantizations like BF16 (24.3 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
What's the best quantization for Gemma 4 12B IT Assistant?
For Gemma 4 12B IT Assistant, Q4_K_M (7.5 GB) offers the best balance of quality and VRAM usage. Q5_K_M (8.9 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 5.4 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Gemma 4 12B IT Assistant on a Mac?
Yes — Mac Mini M2 Pro (16 GB) and 133 other Macs can run Gemma 4 12B IT Assistant. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 7.5 GB of usable unified memory (RAM minus macOS overhead).
Can I run Gemma 4 12B IT Assistant locally?
Yes — Gemma 4 12B IT Assistant can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Gemma 4 12B IT Assistant?
At Q4_K_M, Gemma 4 12B IT Assistant can reach ~638 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~87 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 7.5200 × 0.65 = ~691 tok/s
Estimated speed at Q4_K_M (7.5 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Gemma 4 12B IT Assistant?
At Q4_K_M, the download is about 7.20 GB. The full-precision BF16 version is 24.00 GB. The smallest option (Q2_K) is 5.10 GB.
Which GPUs can run Gemma 4 12B IT Assistant?
52 consumer GPUs can run Gemma 4 12B IT Assistant at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 7600. 36 GPUs have plenty of headroom for comfortable inference.
Which devices can run Gemma 4 12B IT Assistant?
151 devices with unified memory can run Gemma 4 12B IT Assistant at Q4_K_M (7.5 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.