Gemma 4 31B Protokoll — Hardware Requirements & GPU Compatibility
ChatSummary
Gemma 4 31B Protokoll is a 31B-parameter open language model from aihpi in the Gemma 4 family. At BF16 it needs about 68.20 GB of VRAM — see which GPUs and Macs can run it below.
Gemma 4 31B Protokoll runs on a 96 GB+ Mac or a workstation GPU
Best local pick → Mac Studio (M3 Ultra, 96GB) · BF16 · 72 GB · ~8 tok/s
Macs from 96 GB (BF16 · ~8 tok/s)
Gemma Terms
Not set up for it?
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Gemma 4 31B Protokoll
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 aihpi/gemma-4-31b-protokoll --max-model-len 16384 --tensor-parallel-size 8vLLM repo with its own quantisation
Download: Weights on Hugging Face
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX PRO 6000 (96 GB) (24 GB headroom) · BF16
- Generation speed
- ~17 tok/s
- generation speed
- Cost per 1M output tokens
- $21.63
- per 1M output tokens
How Much VRAM Does Gemma 4 31B Protokoll Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | Runs on | |
|---|---|---|---|---|
| 68.2 GB | 76.2 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Runs on RTX PRO 5000 Blackwell 72 GB |
Selected BF16
Brain floating point 1672.0GBat 16K context
- Weights
- 67.4 GB
- KV cache
- 4.3 GB
- Overhead
- 0.3 GB
Server-class: ~72 GB at 16K context · one RTX PRO 5000 Blackwell 72 GB
Or .
Which Hardware Can Run Gemma 4 31B Protokoll?
BF16 · 68.2 GBGemma 4 31B Protokoll (BF16) requires 68.2 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 89+ GB is recommended. No consumer GPU has enough memory. 30 devices with unified memory can run Gemma 4 31B Protokoll, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Studio (M3 Ultra, 96GB).
96 GB
fits · 24.1 GB free · up to 32K ctx
3 GPUs · 2 Mac classes
RTX PRO 6000 Blackwell Max-Q Workstation Edition~17Simulate output on RTX PRO 6000 Blackwell Max-Q Workstation EditionRTX PRO 6000 Blackwell Workstation Edition~17Simulate output on RTX PRO 6000 Blackwell Workstation EditionRTX PRO 6000 Blackwell Server Edition~15Simulate output on RTX PRO 6000 Blackwell Server EditionM3 Ultra*96 GB~8Simulate output on Mac Studio (M3 Ultra, 96GB)M3 Max*96 GB~3Simulate output on MacBook Pro 14" M3 Max (96 GB)128–144 GB
fits · 56.0 GB free · up to 32K ctx
4 GPUs · 2 Mac classes · 4 device families
GH200 Grace Hopper Superchip~47Simulate output on GH200 Grace Hopper SuperchipH200 NVL~46Simulate output on H200 NVLH200 SXM~46Simulate output on H200 SXMM2 Ultra*128 GB~8Simulate output on Mac Pro M2 Ultra (128 GB)M5 Max128 GB~6Simulate output on Mac Studio M5 Max (128 GB)GB10~3Simulate output on ASUS Ascent GX10Jetson T5000 (Thor)~3Simulate output on NVIDIA Jetson AGX Thor Developer Kit
By VRAM tier · typical GPUs and Macs at BF16
Consumer GPUs (up to 32 GB) and Macs up to 64 GB: too small for this model (needs ≥ 72 GB)8 GB tier →12 GB tier →16 GB tier →24 GB tier →Macs →
Alternatives to Gemma 4 31B Protokoll
Gemma 4 family · 35 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Nanbeige4.3 33BNanbeige | — | 23.7 GB | ~32 | same size |
| Hy Memory 30B A3BTencent | — | 19.1 GB | ~184 | same size |
| Granite Switch 4.1 30BIBM | — | 22.7 GB | ~35 | same size |
| Llm Jp 4.1 32B A3b Thinkingllm-jp | — | 20.1 GB | ~188 | same size |
| Llm Jp 4.1 33B Thinkingllm-jp | — | 22.5 GB | ~35 | same size |
| Agnes 3.0 FlashAgnes-AI | — | 22.1 GB | ~35 | same size |
Specifications
- Publisher
- aihpi
- Family
- Gemma 4
- Parameters
- 31B
- Release Date
- 2026-09-17
- License
- Gemma Terms
Frequently Asked Questions
How much VRAM does Gemma 4 31B Protokoll need?
Gemma 4 31B Protokoll requires 68.2 GB of VRAM at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 31B × 16 bits ÷ 8 = 62 GB
KV Cache + Overhead ≈ 6.2 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.
VRAM usage by quantization
Can NVIDIA GeForce RTX 5090 run Gemma 4 31B Protokoll?
No — Gemma 4 31B Protokoll requires at least 68.2 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
Can I run Gemma 4 31B Protokoll on a Mac?
Yes — Mac Studio (M3 Ultra, 96GB) and 20 other Macs can run Gemma 4 31B Protokoll. Apple Silicon uses unified memory, so the model shares RAM with the system. At BF16 you need at least 68.2 GB of usable unified memory (RAM minus macOS overhead).
Can I run Gemma 4 31B Protokoll locally?
Yes — Gemma 4 31B Protokoll can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Gemma 4 31B Protokoll?
At BF16, Gemma 4 31B Protokoll can reach ~70 tok/s on AMD Instinct MI350X. 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 ÷ 68.2 × 0.65 = ~76 tok/s
Estimated speed at BF16 (68.2 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 31B Protokoll?
At BF16, the download is about 62.00 GB.
Which GPUs can run Gemma 4 31B Protokoll?
No single consumer GPU has enough VRAM to run Gemma 4 31B Protokoll at BF16 (68.2 GB). Multi-GPU or professional hardware is required.
Which devices can run Gemma 4 31B Protokoll?
30 devices with unified memory can run Gemma 4 31B Protokoll at BF16 (68.2 GB), including ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB), GMKtec EVO-X2 (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.