Gemma 3 12B IT GGUF — Hardware Requirements & GPU Compatibility
Vision
105.1K downloads
157 likesSpecifications
- Publisher
- Unsloth
- Family
- Gemma
- Parameters
- 12B
- Architecture
- Gemma3ForConditionalGeneration
- Context Length
- 131,072 tokens
- Vocabulary Size
- 262,208
- License
- Gemma Terms
Get Started
HuggingFace
How Much VRAM Does Gemma 3 12B IT GGUF Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| IQ2_XXS | 2.20 | 4.3 GB | 51.9 GB | 3.30 GB | Importance-weighted 2-bit, extreme compression — significant quality loss |
| IQ2_M | 2.70 | 5.1 GB | 52.7 GB | 4.05 GB | Importance-weighted 2-bit, medium |
| IQ3_XXS | 3.10 | 5.7 GB | 53.3 GB | 4.65 GB | Importance-weighted 3-bit |
| Q2_K | 3.40 | 6.2 GB | 53.7 GB | 5.10 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 6.3 GB | 53.9 GB | 5.25 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 6.9 GB | 54.5 GB | 5.85 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 7.0 GB | 54.6 GB | 6.00 GB | 4-bit legacy quantization |
| IQ4_XS | 4.30 | 7.5 GB | 55.1 GB | 6.45 GB | Importance-weighted 4-bit, compact |
| Q4_1 | 4.50 | 7.8 GB | 55.4 GB | 6.75 GB | 4-bit legacy quantization with offset |
| Q4_K_S | 4.50 | 7.8 GB | 55.4 GB | 6.75 GB | 4-bit small quantization |
| IQ4_NL | 4.50 | 7.8 GB | 55.4 GB | 6.75 GB | Importance-weighted 4-bit, non-linear |
| Q4_K_M | 4.80 | 8.3 GB | 55.8 GB | 7.20 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_S | 5.50 | 9.3 GB | 56.9 GB | 8.25 GB | 5-bit small quantization |
| Q5_K_M | 5.70 | 9.6 GB | 57.2 GB | 8.55 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 10.9 GB | 58.5 GB | 9.90 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 13.1 GB | 60.6 GB | 12.00 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run Gemma 3 12B IT GGUF?
Q4_K_M · 8.3 GBGemma 3 12B IT GGUF (Q4_K_M) requires 8.3 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 11+ GB is recommended. Using the full 131K context window can add up to 47.6 GB, bringing total usage to 55.8 GB. 28 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3080 Ti.
Runs great
— Plenty of headroom
Decent
— Enough VRAM, may be tightWhich Devices Can Run Gemma 3 12B IT GGUF?
Q4_K_M · 8.3 GB27 devices with unified memory can run Gemma 3 12B IT GGUF, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomRelated Models
Frequently Asked Questions
- How much VRAM does Gemma 3 12B IT GGUF need?
Gemma 3 12B IT GGUF requires 8.3 GB of VRAM at Q4_K_M, or 13.1 GB at Q8_0. Full 131K context adds up to 47.6 GB (55.8 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 12B × 4.8 bits ÷ 8 = 7.2 GB
KV Cache + Overhead ≈ 1 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 48.6 GB (at full 131K context)
VRAM usage by quantization
Q4_K_M8.3 GBQ4_K_M + full context55.8 GB- What's the best quantization for Gemma 3 12B IT GGUF?
For Gemma 3 12B IT GGUF, Q4_K_M (8.3 GB) offers the best balance of quality and VRAM usage. Q5_K_S (9.3 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 4.3 GB.
VRAM requirement by quantization
IQ2_XXS4.3 GB~53%Q3_K_S6.3 GB~77%Q4_17.8 GB~88%Q4_K_M ★8.3 GB~89%Q5_K_S9.3 GB~92%Q8_013.1 GB~99%★ Recommended — best balance of quality and VRAM usage.
- Can I run Gemma 3 12B IT GGUF on a Mac?
Gemma 3 12B IT GGUF requires at least 4.3 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 GGUF locally?
Yes — Gemma 3 12B IT GGUF can run locally on consumer hardware. At Q4_K_M quantization it needs 8.3 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Gemma 3 12B IT GGUF?
At Q4_K_M, Gemma 3 12B IT GGUF can reach ~353 tok/s on AMD Instinct MI300X. On NVIDIA GeForce RTX 4090: ~79 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 ÷ 8.3 × 0.55 = ~353 tok/s
Estimated speed at Q4_K_M (8.3 GB)
AMD Instinct MI300X~353 tok/sNVIDIA GeForce RTX 4090~79 tok/sNVIDIA H100 SXM~264 tok/sAMD Instinct MI250X~219 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 GGUF?
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.