Gemma 3 12B IT — Hardware Requirements & GPU Compatibility
Vision
Google Gemma 3 12B IT is a 12-billion parameter multimodal instruction-tuned model from Google's Gemma 3 series. It supports both text and image inputs, offering vision-language capabilities at a more accessible size point than the 27B variant. Gemma 3 12B IT runs on consumer GPUs with 12-16GB of VRAM in quantized formats, making it a practical choice for local multimodal inference without requiring top-tier hardware. Released under the Gemma license.
1.9M downloads
678 likesSpecifications
- Publisher
- Family
- Gemma
- Parameters
- 12B
- Context Length
- 32,768 tokens
- License
- Gemma Terms
Get Started
HuggingFace
How Much VRAM Does Gemma 3 12B IT Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| IQ2_XXS | 2.20 | 3.6 GB | — | 3.30 GB | Importance-weighted 2-bit, extreme compression — significant quality loss |
| IQ2_M | 2.70 | 4.5 GB | — | 4.05 GB | Importance-weighted 2-bit, medium |
| IQ3_XXS | 3.10 | 5.1 GB | — | 4.65 GB | Importance-weighted 3-bit |
| IQ3_XS | 3.30 | 5.5 GB | — | 4.95 GB | Importance-weighted 3-bit, extra small |
| IQ3_S | 3.40 | 5.6 GB | — | 5.10 GB | Importance-weighted 3-bit, small |
| Q2_K | 3.40 | 5.6 GB | — | 5.10 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 5.8 GB | — | 5.25 GB | 3-bit small quantization |
| IQ3_M | 3.60 | 5.9 GB | — | 5.40 GB | Importance-weighted 3-bit, medium |
| Q3_K_M | 3.90 | 6.4 GB | — | 5.85 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 6.6 GB | — | 6.00 GB | 4-bit legacy quantization |
| Q3_K_L | 4.10 | 6.8 GB | — | 6.15 GB | 3-bit large quantization |
| IQ4_XS | 4.30 | 7.1 GB | — | 6.45 GB | Importance-weighted 4-bit, compact |
| Q4_1 | 4.50 | 7.4 GB | — | 6.75 GB | 4-bit legacy quantization with offset |
| Q4_K_S | 4.50 | 7.4 GB | — | 6.75 GB | 4-bit small quantization |
| IQ4_NL | 4.50 | 7.4 GB | — | 6.75 GB | Importance-weighted 4-bit, non-linear |
| Q4_K_M | 4.80 | 7.9 GB | — | 7.20 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_S | 5.50 | 9.1 GB | — | 8.25 GB | 5-bit small quantization |
| Q5_K_M | 5.70 | 9.4 GB | — | 8.55 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 10.9 GB | — | 9.90 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 13.2 GB | — | 12.00 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run Gemma 3 12B IT?
Q4_K_M · 7.9 GBGemma 3 12B IT (Q4_K_M) requires 7.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 11+ GB is recommended. 35 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3080.
Runs great
— Plenty of headroom
Decent
— Enough VRAM, may be tightWhich Devices Can Run Gemma 3 12B IT?
Q4_K_M · 7.9 GB33 devices with unified memory can run Gemma 3 12B IT, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, MacBook Air 13" M3 (8 GB).
Runs great
— Plenty of headroomDecent
— Enough memory, may be tightRelated Models
Derivatives (8)
Frequently Asked Questions
- How much VRAM does Gemma 3 12B IT need?
Gemma 3 12B IT requires 7.9 GB of VRAM at Q4_K_M, or 13.2 GB at Q8_0.
VRAM = Weights + KV Cache + Overhead
Weights = 12B × 4.8 bits ÷ 8 = 7.2 GB
KV Cache + Overhead ≈ 0.7 GB (at 2K context + ~0.3 GB framework)
VRAM usage by quantization
Q4_K_M7.9 GB- What's the best quantization for Gemma 3 12B IT?
For Gemma 3 12B IT, Q4_K_M (7.9 GB) offers the best balance of quality and VRAM usage. Q5_K_S (9.1 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 3.6 GB.
VRAM requirement by quantization
IQ2_XXS3.6 GB~53%Q2_K5.6 GB~75%Q3_K_L6.8 GB~86%IQ4_NL7.4 GB~88%Q4_K_M ★7.9 GB~89%Q8_013.2 GB~99%★ Recommended — best balance of quality and VRAM usage.
- Can I run Gemma 3 12B IT on a Mac?
Gemma 3 12B IT requires at least 3.6 GB at IQ2_XXS, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Gemma 3 12B IT locally?
Yes — Gemma 3 12B IT can run locally on consumer hardware. At Q4_K_M quantization it needs 7.9 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Gemma 3 12B IT?
At Q4_K_M, Gemma 3 12B IT can reach ~368 tok/s on AMD Instinct MI300X. On NVIDIA GeForce RTX 4090: ~83 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: AMD Instinct MI300X → 5300 ÷ 7.9 × 0.55 = ~368 tok/s
Estimated speed at Q4_K_M (7.9 GB)
AMD Instinct MI300X~368 tok/sNVIDIA GeForce RTX 4090~83 tok/sNVIDIA H100 SXM~275 tok/sAMD Instinct MI250X~228 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Gemma 3 12B IT?
At Q4_K_M, the download is about 7.20 GB. The full-precision Q8_0 version is 12.00 GB. The smallest option (IQ2_XXS) is 3.30 GB.