SuperGemma 4 12B Abliterated — Hardware Requirements & GPU Compatibility
ChatCode
SuperGemma 4 12B Abliterated is a 12.0B-parameter open language model from Jiunsong in the Gemma 4 family. It supports a context window of up to 262,144 tokens. At Q4_K_M it needs about 8.23 GB of VRAM — see which GPUs and Macs can run it below.
288 downloads
10 likesSpecifications
- Publisher
- Jiunsong
- Family
- Gemma 4
- Parameters
- 12.0B
- Architecture
- Gemma4UnifiedForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 262,144
- Release Date
- 2026-06-25
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does SuperGemma 4 12B Abliterated Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_Kest. | 3.40 | 6.1 GB | 102.0 GB | 5.08 GB | 2-bit quantization with K-quant improvements |
| Q3_K_Mest. | 3.90 | 6.9 GB | 102.8 GB | 5.83 GB | 3-bit medium quantization |
| Q4_K_M | 4.80 | 8.2 GB | 104.1 GB | 7.18 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_Mest. | 5.70 | 9.6 GB | 105.5 GB | 8.52 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_Kest. | 6.60 | 10.9 GB | 106.8 GB | 9.87 GB | 6-bit quantization, very good quality |
| Q8_0est. | 8.00 | 13.0 GB | 108.9 GB | 11.96 GB | 8-bit quantization, near-lossless |
| BF16est. | 16.00 | 25.0 GB | 120.9 GB | 23.92 GB | Brain floating point 16 — preferred for training |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run SuperGemma 4 12B Abliterated?
Q4_K_M · 8.2 GBSuperGemma 4 12B Abliterated (Q4_K_M) requires 8.2 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 262K context window can add up to 95.9 GB, bringing total usage to 104.1 GB. 39 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 SuperGemma 4 12B Abliterated?
Q4_K_M · 8.2 GB49 devices with unified memory can run SuperGemma 4 12B Abliterated, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, iPad Pro M5 13" (16 GB).
Runs great
— Plenty of headroomDecent
— Enough memory, may be tightWhere to Download SuperGemma 4 12B Abliterated
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Related Models
Frequently Asked Questions
- How much VRAM does SuperGemma 4 12B Abliterated need?
SuperGemma 4 12B Abliterated requires 8.2 GB of VRAM at Q4_K_M, or 25.0 GB at BF16. Full 262K context adds up to 95.9 GB (104.1 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 12.0B × 4.8 bits ÷ 8 = 7.2 GB
KV Cache + Overhead ≈ 1 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 96.9 GB (at full 262K context)
VRAM usage by quantization
Q4_K_M8.2 GBQ4_K_M + full context104.1 GB- Can NVIDIA GeForce RTX 4090 run SuperGemma 4 12B Abliterated?
Yes, at Q8_0 (13.0 GB) or lower. Higher quantizations like BF16 (25.0 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
- What's the best quantization for SuperGemma 4 12B Abliterated?
For SuperGemma 4 12B Abliterated, Q4_K_M (8.2 GB) offers the best balance of quality and VRAM usage. Q5_K_M (9.6 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 6.1 GB.
VRAM requirement by quantization
Q2_K6.1 GBQ4_K_M ★8.2 GBQ5_K_M9.6 GBQ6_K10.9 GBQ8_013.0 GBBF1625.0 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run SuperGemma 4 12B Abliterated on a Mac?
SuperGemma 4 12B Abliterated requires at least 6.1 GB at Q2_K, 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 SuperGemma 4 12B Abliterated locally?
Yes — SuperGemma 4 12B Abliterated can run locally on consumer hardware. At Q4_K_M quantization it needs 8.2 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is SuperGemma 4 12B Abliterated?
At Q4_K_M, SuperGemma 4 12B Abliterated can reach ~535 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~80 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 ÷ 8.2 × 0.65 = ~632 tok/s
Estimated speed at Q4_K_M (8.2 GB)
~632 tok/s~80 tok/s~632 tok/s~535 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of SuperGemma 4 12B Abliterated?
At Q4_K_M, the download is about 7.18 GB. The full-precision BF16 version is 23.92 GB. The smallest option (Q2_K) is 5.08 GB.
- Which GPUs can run SuperGemma 4 12B Abliterated?
39 consumer GPUs can run SuperGemma 4 12B Abliterated at Q4_K_M (8.2 GB). Top options include AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 6900 XT, AMD Radeon RX 6700 XT. 26 GPUs have plenty of headroom for comfortable inference.
- Which devices can run SuperGemma 4 12B Abliterated?
52 devices with unified memory can run SuperGemma 4 12B Abliterated at Q4_K_M (8.2 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Apple iPhone 17 Pro, Asus ROG Flow Z13 (2025, 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.