Expired50% confidenceFactExact time
After quantization, Gemma 4 31B has a model size of approximately 20GB, with measured GPU memory usage of approximately 20.5GB and GPU utilization of approximately 91%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Gemma 4 Open-Source Model Local Deployment Guide: Ollama Installation & Mobile Setup
bilibili搞机零距离-6/10/2026
Related Claims
UnverifiedThe Gemma 4 26B model is approximately 17 GB and is best suited for GPUs with 24 GB or more of VRAM, delivering the best programming performance79% similarUnverifiedThe Gemma 4 26B MOE version achieves an actual inference speed of 60 tokens per second on a GPU with 24GB of VRAM.78% similarUnverifiedThe model weights of Flux 2's 9B parameter model occupy approximately 18GB of VRAM at FP16 precision, typically requiring more than 24GB of GPU VRAM.78% similarUnverifiedThe FP16 version of Gemma 3 12B requires approximately 24 GB of VRAM, while the Q4_K_M version reduces this to approximately 7–8 GB, with inference speed improving by approximately 2–3x and structured output accuracy losing only approximately 2–5%.77% similarUnverifiedConsumer-grade GPUs such as the RTX 4090 have 24 GB of VRAM and can run quantized models with billions of parameters.76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/47981API
curl https://kongchang.com/api/v1/knowledge/claims/47981MCP
get_claim(id=47981)