FitMyLLM
Google/Dense

GoogleGemma 4 31B

Gemma 4 31B — Google's most capable open model. Dense 31B with 256K context, vision, and thinking mode.

chatcodingreasoningmultilingualvision
32.68B
Parameters
256K
Context length
19
Benchmarks
14
Quantizations
500K
HF downloads
Architecture
Dense
Released
2026-04-02
Layers
60
KV Heads
16
Head Dim
256
Family
gemma

Quantization Options

Context length:
QuantBitsVRAM @ 16KQuality
IQ3_XXS3.25
16.4 GB
13.8 + 2.7 KV
low
IQ3_XS3.5
17.4 GB
14.8 + 2.7 KV
low
Q3_K_S3.64
18.0 GB
15.4 + 2.7 KV
low
IQ3_M3.76
18.5 GB
15.8 + 2.7 KV
low
Q3_K_M4
19.5 GB
16.8 + 2.7 KV
low
Q3_K_L4.3
20.7 GB
18.1 + 2.7 KV
moderate
IQ4_XS4.46
21.4 GB
18.7 + 2.7 KV
moderate
Q4_K_S4.67
22.2 GB
19.6 + 2.7 KV
moderate
Q4_K_M4.89
23.1 GB
20.5 + 2.7 KV
good
Q5_K_S5.57
25.9 GB
23.2 + 2.7 KV
good
Q5_K_M5.7
26.4 GB
23.8 + 2.7 KV
good
Q6_K6.56
29.9 GB
27.3 + 2.7 KV
excellent
Q8_08.5
37.9 GB
35.2 + 2.7 KV
lossless
FP1616
68.5 GB
65.8 + 2.7 KV
lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

Deploying for a team or in production? Size GPUs, cost & scaling in Enterprise →
▸ READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN GEMMA 4 31B NOW

Rent on Vast.ai →Or RunPod →

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (19)

Arena Elo1452
AIME89.2
MMLU-PRO85.2
GPQA Diamond84.3
LiveCodeBench80.0
IFBench75.6
IFEval75.5
BBH74.4
AA Long Context62.0
τ²-Bench59.9
SciCode43.4
BigCodeBench42.8
AA Intelligence39.2
AA Coding38.7
Terminal-Bench36.4
MATH27.9
HLE19.5
MUSR16.9
GPQA16.0

Run this model

▸Easiest way to get started·Beginners
DOCS ↗
curl -fsSL https://ollama.com/install.sh | sh
$ollama run gemma4:31b-it-q4_K_M

Downloads and runs automatically. Add --verbose for speed stats.

▸ SETUP GUIDE
>_

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more
Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines
HuggingFaceOllama LibraryGGUF DownloadsBuild Hardware

GPUs that can run this model

At Q4_K_M quantization. Sorted by minimum VRAM.

NVIDIA RTX 4090
24 GB VRAM • 1008 GB/s
NVIDIA
$1599
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA RTX 3090 Ti
24 GB VRAM • 1008 GB/s
NVIDIA
$999
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA RTX 3090
24 GB VRAM • 936 GB/s
NVIDIA
$850
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
AMD RX 7900 XTX
24 GB VRAM • 960 GB/s
AMD
$999
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
Apple M4 Pro (24GB)
24 GB VRAM • 273 GB/s
APPLE
$1399
NVIDIA L4 24GB
24 GB VRAM • 300 GB/s
NVIDIA
$2500
NVIDIA A10 24GB
24 GB VRAM • 600 GB/s
NVIDIA
$3500
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
Apple M2 (24GB)
24 GB VRAM • 100 GB/s
APPLE
$999
Apple M3 (24GB)
24 GB VRAM • 100 GB/s
APPLE
$999
Apple M4 (24GB)
24 GB VRAM • 120 GB/s
APPLE
$699
NVIDIA Tesla M40 24 GB
24 GB VRAM • 288 GB/s
NVIDIA
NVIDIA Tesla P10
24 GB VRAM • 694 GB/s
NVIDIA
NVIDIA Tesla P40
24 GB VRAM • 347 GB/s
NVIDIA
NVIDIA Quadro RTX 6000
24 GB VRAM • 672 GB/s
NVIDIA
$4000
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA Quadro RTX 6000 Passive
24 GB VRAM • 624 GB/s
NVIDIA
$4000
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA GeForce RTX 3090
24 GB VRAM • 936 GB/s
NVIDIA
$1499
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA A10 PCIe
24 GB VRAM • 600 GB/s
NVIDIA
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA A10G
24 GB VRAM • 600 GB/s
NVIDIA
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA RTX A5000
24 GB VRAM • 768 GB/s
NVIDIA
$2500
NVIDIA GeForce RTX 3090 Ti
24 GB VRAM • 1010 GB/s
NVIDIA
$1999
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA GeForce RTX 4090
24 GB VRAM • 1010 GB/s
NVIDIA
$1599
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA L40 CNX
24 GB VRAM • 864 GB/s
NVIDIA
$5000
NVIDIA L40G
24 GB VRAM • 864 GB/s
NVIDIA
$5000
AMD Radeon RX 7900 XTX
24 GB VRAM • 960 GB/s
AMD
$999
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA GeForce RTX 4090 D
24 GB VRAM • 1010 GB/s
NVIDIA
$1599
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA GeForce RTX 5090 D V2
24 GB VRAM • 1340 GB/s
NVIDIA
$1999
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA TITAN RTX
24 GB VRAM • 672 GB/s
NVIDIA
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA A30 PCIe
24 GB VRAM • 933 GB/s
NVIDIA
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA A30X
24 GB VRAM • 1220 GB/s
NVIDIA
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai
NVIDIA PG506-207
24 GB VRAM • 933 GB/s
NVIDIA
AmazonNeweggeBayPCPartPickerRent on RunPodRent on Vast.ai

Find the best GPU for Gemma 4 31B

Build Hardware for Gemma 4 31B
▸ SPEC SHEET

Gemma 4 31B — 32.68B Dense.

▸ SPECIFICATIONS
PARAMETERS
32.68B
ARCHITECTURE
Dense Transformer
CONTEXT LENGTH
256K tokens
CAPABILITIES
chat, coding, reasoning, multilingual, vision
RELEASE DATE
2026-04-02
PROVIDER
Google
FAMILY
gemma
▸ VRAM REQUIREMENTS
QUANTBPWVRAMQUALITY
IQ3_XXS3.2513.8 GB82%
IQ3_XS3.514.8 GB84%
Q3_K_S3.6415.4 GB85%
IQ3_M3.7615.8 GB86%
Q3_K_M416.8 GB88%
Q3_K_L4.318.1 GB90%
IQ4_XS4.4618.7 GB92%
Q4_K_S4.6719.6 GB93%
Q4_K_M4.8920.5 GB94%
Q5_K_S5.5723.2 GB96%
Q5_K_M5.723.8 GB96%
Q6_K6.5627.3 GB97%
Q8_08.535.2 GB100%
FP161665.8 GB100%
§ 01BENCHMARK SCORES
MMLU-PRO85.2
MATH27.9
IFEval75.5
BBH74.4
GPQA16.0
MUSR16.9
BigCodeBench42.8
Arena Elo1452.0
GPQA Diamond84.3
HLE19.5
AA Intelligence39.2
AA Coding38.7
LiveCodeBench80.0
AIME89.2
aa_ifbench75.6
aa_terminal_bench36.4
aa_tau259.9
aa_scicode43.4
aa_lcr62.0
§ 02RUN COMMAND

Run Gemma 4 31B locally with Ollama — needs 20.5 GB VRAM at Q4_K_M:

$ollama run gemma4:31b
§ 03COMPATIBLE GPUs
30 @ Q4_K_M
NVIDIA RTX 4090
24 GB · 1008 GB/s
NVIDIA RTX 3090 Ti
24 GB · 1008 GB/s
NVIDIA RTX 3090
24 GB · 936 GB/s
AMD RX 7900 XTX
24 GB · 960 GB/s
Apple M4 Pro (24GB)
24 GB · 273 GB/s
NVIDIA L4 24GB
24 GB · 300 GB/s
NVIDIA A10 24GB
24 GB · 600 GB/s
Apple M2 (24GB)
24 GB · 100 GB/s
Apple M3 (24GB)
24 GB · 100 GB/s
Apple M4 (24GB)
24 GB · 120 GB/s
NVIDIA Tesla M40 24 GB
24 GB · 288 GB/s
NVIDIA Tesla P10
24 GB · 694 GB/s
NVIDIA Tesla P40
24 GB · 347 GB/s
NVIDIA Quadro RTX 6000
24 GB · 672 GB/s
NVIDIA Quadro RTX 6000 Passive
24 GB · 624 GB/s