CanItRun Logocanitrun.

Gemma 3 1B Instruct

Gemma 3 1B Instruct needs roughly 1.1 GB VRAM at Q4_K_M quantization (2.6 GB at FP16). 117 GPUs we track can run it fully in VRAM at 8k context.

117 GPUs run this natively · 1 with CPU offload

Google1B params32k contextGemmaCommercial use ok

Gemma 3 1B Instruct is a 1B parameter dense model developed by Google. March 2025 release, the smallest model in the Gemma 3 family. Unlike the 4B/12B/27B siblings, the 1B checkpoint is text-only (no SigLIP vision encoder) and caps out at 32K context rather than 128K.

To run Gemma 3 1B Instruct locally: Q8_0 needs roughly 1GB and runs on any GPU, most phones, and even Raspberry Pi-class hardware without issue.

MMLU-Pro 14.7 is modest, as expected at 1B scale, but the tradeoff buys very low latency for simple classification, autocomplete, and on-device assistant tasks.

VRAM at each quantization

Calculated at 8k context. Since KV cache scales linearly with context, longer sessions need more VRAM than shown here.

QuantWeightsKV cacheTotal
FP324.0 GB0.33 GB4.8 GB
BF162.0 GB0.33 GB2.6 GB
FP162.0 GB0.33 GB2.6 GB
Q8_0rec1.1 GB0.33 GB1.6 GB
Q6_K0.8 GB0.33 GB1.3 GB
Q5_K_M0.7 GB0.33 GB1.2 GB
Q4_K_M0.6 GB0.33 GB1.1 GB
Q3_K_M0.5 GB0.33 GB0.9 GB
Q2_K0.4 GB0.33 GB0.8 GB
NVFP4cuda0.5 GB0.33 GB0.9 GB

Shown at 8k context with FP16 KV cache. NVFP4 needs a CUDA GPU to run. Toggle TurboQuant in the calculator to view compressed KV cache numbers.

Benchmarks

MMLU-Pro
14.7

GPUs that run Gemma 3 1B Instruct natively (117)

Show 112 more
Plus 1 GPU that runs it with CPU offload (slower)

No GPU? Rent an RTX 3080 10GB from $0.17/hr ↗

Prices as of Oct 17, 2026. Rates change; check RunPod for current pricing. Affiliate link: CanItRun may earn a commission. Doesn't affect the fit calculation above.

Notes

Text-only; larger siblings support vision.

Hugging Face ↗Ollama ↗Released 2025-03-12

Compare Gemma 3 1B Instruct with other models

Explore

Browse all guides →Browse all models →Browse all GPUs →

Frequently asked questions

What are the VRAM requirements for Gemma 3 1B Instruct?
Gemma 3 1B Instruct requires approximately 1.1 GB of VRAM at Q4_K_M quantization, 1.6 GB at Q8, and 2.6 GB at FP16. These numbers assume 8k context window; VRAM scales linearly with context length due to the KV cache.
How many parameters does Gemma 3 1B Instruct have?
Gemma 3 1B Instruct has 1 billion parameters.
How capable is Gemma 3 1B Instruct?
Gemma 3 1B Instruct has an MMLU-Pro score of 14.7, making it well-suited for lightweight tasks, prototyping, and resource-constrained environments.
Can Gemma 3 1B Instruct run on a 16 GB GPU?
Yes. Gemma 3 1B Instruct needs 1.1 GB at Q4_K_M, which fits in a 16 GB GPU like the RTX 4080 or RTX 5070 Ti.
What is the smallest quantization for Gemma 3 1B Instruct that fits in 24 GB of VRAM?
At BF16, Gemma 3 1B Instruct needs 2.6 GB, the highest-quality quantization that fits in 24 GB of VRAM.
What GPU do I need to run Gemma 3 1B Instruct locally?
A 16 GB GPU is enough. At Q4_K_M, Gemma 3 1B Instruct needs 1.1 GB VRAM. Good options: RTX 4080 (16 GB), RTX 5070 Ti (16 GB).
Can I run Gemma 3 1B Instruct in the cloud?
Yes. The cheapest RunPod setup we found that holds Gemma 3 1B Instruct entirely in GPU memory at 8k context is 1× RTX 3080 10GB at BF16, at roughly 198.4 tokens per second and from $0.17 per hour for the whole pod. Prices as of Oct 17, 2026; RunPod rates change, so check their pricing page for the current figure.