How much VRAM does Granite 3.1 MoE 3B Q8 need?

Granite 3.1 MoE 3B Q8 needs about 10GB minimum VRAM and 20GB optimal VRAM for a safer local run target.

Can Granite 3.1 MoE 3B Q8 run on an RTX 3090 24GB?

Yes. Granite 3.1 MoE 3B Q8 fits comfortably on an RTX 3090 24GB target with 20GB optimal VRAM.

What is the Ollama command for Granite 3.1 MoE 3B Q8?

Use ollama run granite3.1-moe:3b. Check the Ollama tag granite3.1-moe:3b before running.

What is the approximate Ollama model size for Granite 3.1 MoE 3B Q8?

Granite 3.1 MoE 3B Q8 uses the granite3.1-moe:3b Ollama tag. Treat the Q8 3 profile as smaller than the runtime VRAM budget; use 10GB minimum VRAM and 20GB optimal VRAM as the safer sizing numbers.

How much system RAM does Granite 3.1 MoE 3B Q8 need in Ollama?

Plan for at least 32GB system RAM alongside 10GB minimum VRAM and 20GB optimal VRAM. More RAM helps if layers spill to CPU or if you run multiple models.

What is the RAM vs VRAM requirement for Granite 3.1 MoE 3B Q8?

Granite 3.1 MoE 3B Q8 should be sized by GPU VRAM first: 10GB minimum and 20GB optimal. System RAM should be at least 32GB so Ollama has room for CPU-side layers, context, and the surrounding app process.

What is Granite 3.1 MoE 3B Q8 VRAM usage on an RTX 3090?

Granite 3.1 MoE 3B Q8 is comfortable for an RTX 3090 24GB target. The page estimates 10GB minimum and 20GB optimal VRAM, with spill risk marked as moderate at long context.

Which LocalVRAM profiles share the granite3.1-moe:3b Ollama tag?

LocalVRAM tracks these profiles for the same Ollama tag or model family: Granite 3.1 MoE 3B Q8 (Q8, 10GB min/20GB optimal); Granite 3.1 MoE 3B Q4 (Q4, 4GB min/14GB optimal); Granite 3.1 MoE 3B Q5 (Q5, 6GB min/16GB optimal); Granite 3.1 MoE 3B FP16 (FP16, 16GB min/28GB optimal). Use the table on this page to compare quantization and VRAM targets before choosing a run command.

Is Granite 3.1 MoE 3B Q8 benchmark data measured or estimated?

Granite 3.1 MoE 3B Q8 currently uses estimated baseline anchors on this page. Verified benchmark data will replace the estimate after a local hardware run is available.

Granite 3.1 MoE 3B Q8 VRAM Requirement and Ollama Model Size

Granite 3.1 MoE 3B Q8 needs 10GB minimum VRAM, 20GB optimal VRAM, and 32GB+ system RAM for a safer Ollama run. Popular Ollama model family: Granite 3.1 MoE. Caveat: Estimated values are placeholders unless marked measured..

Hardware Snapshot

Family	Granite 3.1 MoE
Scenario	reasoning
License scope	open-source
Quantization	Q8
VRAM minimum	10GB
VRAM optimal	20GB
Best local GPU	RTX 3090 24GB
Cloud fallback	A6000 48GB
Updated	2026-02-24
Data status	Estimated baseline (pending measurement)
Ollama source	Library reference (verified: 2026-02-24)
Ollama tag	`granite3.1-moe:3b`
Category	reasoning

Benchmark Anchors

Hardware	Expected tok/s
RTX 3090 24GB	25.9
RTX 4090 24GB	35
A100 80GB	62.2

Quick Answers

How much VRAM does Granite 3.1 MoE 3B Q8 need?: Granite 3.1 MoE 3B Q8 needs about 10GB minimum VRAM and 20GB optimal VRAM for a safer local run target.
Can Granite 3.1 MoE 3B Q8 run on an RTX 3090 24GB?: Yes. Granite 3.1 MoE 3B Q8 fits comfortably on an RTX 3090 24GB target with 20GB optimal VRAM.
What is the Ollama command for Granite 3.1 MoE 3B Q8?: Use ollama run granite3.1-moe:3b. Check the Ollama tag granite3.1-moe:3b before running.
What is the approximate Ollama model size for Granite 3.1 MoE 3B Q8?: Granite 3.1 MoE 3B Q8 uses the granite3.1-moe:3b Ollama tag. Treat the Q8 3 profile as smaller than the runtime VRAM budget; use 10GB minimum VRAM and 20GB optimal VRAM as the safer sizing numbers.
How much system RAM does Granite 3.1 MoE 3B Q8 need in Ollama?: Plan for at least 32GB system RAM alongside 10GB minimum VRAM and 20GB optimal VRAM. More RAM helps if layers spill to CPU or if you run multiple models.
What is the RAM vs VRAM requirement for Granite 3.1 MoE 3B Q8?: Granite 3.1 MoE 3B Q8 should be sized by GPU VRAM first: 10GB minimum and 20GB optimal. System RAM should be at least 32GB so Ollama has room for CPU-side layers, context, and the surrounding app process.
What is Granite 3.1 MoE 3B Q8 VRAM usage on an RTX 3090?: Granite 3.1 MoE 3B Q8 is comfortable for an RTX 3090 24GB target. The page estimates 10GB minimum and 20GB optimal VRAM, with spill risk marked as moderate at long context.
Which LocalVRAM profiles share the granite3.1-moe:3b Ollama tag?: LocalVRAM tracks these profiles for the same Ollama tag or model family: Granite 3.1 MoE 3B Q8 (Q8, 10GB min/20GB optimal); Granite 3.1 MoE 3B Q4 (Q4, 4GB min/14GB optimal); Granite 3.1 MoE 3B Q5 (Q5, 6GB min/16GB optimal); Granite 3.1 MoE 3B FP16 (FP16, 16GB min/28GB optimal). Use the table on this page to compare quantization and VRAM targets before choosing a run command.
Is Granite 3.1 MoE 3B Q8 benchmark data measured or estimated?: Granite 3.1 MoE 3B Q8 currently uses estimated baseline anchors on this page. Verified benchmark data will replace the estimate after a local hardware run is available.

Ollama Model Size and RAM Notes

Ollama tag	`granite3.1-moe:3b`
Run command	`ollama run granite3.1-moe:3b`
Model size guidance	Q8 quantization, 3 profile; size runtime by VRAM, not download size alone
Runtime VRAM target	10GB minimum, 20GB optimal
System RAM planning	32GB or more when running this tag locally with Ollama

Download size alone is not a safe runtime budget. Runtime memory also includes KV cache, context length, and GPU layer placement.

Ollama RAM vs VRAM Requirements

GPU VRAM usage estimate	10GB minimum / 20GB optimal
System RAM requirement	32GB or more for Ollama runtime headroom
RTX 3090 24GB fit	Comfortable
CPU spill risk	Moderate at long context
Context warning	Longer context increases KV cache memory, so real VRAM usage can exceed short-prompt estimates.

For Ollama, VRAM is the first fit check. System RAM matters when layers spill to CPU, when context grows, or when the local app stack runs beside the model.

Available Tag Variants and VRAM Targets

Profile	Quantization	VRAM target	Ollama tag
Granite 3.1 MoE 3B Q8	Q8	10GB min / 20GB optimal	`granite3.1-moe:3b`
Granite 3.1 MoE 3B Q4	Q4	4GB min / 14GB optimal	`granite3.1-moe:3b`
Granite 3.1 MoE 3B Q5	Q5	6GB min / 16GB optimal	`granite3.1-moe:3b`
Granite 3.1 MoE 3B FP16	FP16	16GB min / 28GB optimal	`granite3.1-moe:3b`

Ollama library tags often group multiple quantization choices under one model family. LocalVRAM separates them into VRAM profiles so you can pick the right local target.

Real Hardware Benchmark (RTX 3090)

Real benchmark data not available yet for this tag. Estimated anchors are shown above.

Performance Curve

Reference anchors are baseline estimates. Measured RTX 3090 data is overlaid when available.

Best Hardware for Granite 3.1 MoE 3B Q8

Local run: RTX 3090 (24GB) (Check latest deal) for around 25.9 tok/s on this profile.
Cloud run: RunPod A6000 48GB , about 2.4x the local 3090 speed anchor.
Alternative cloud: Vast.ai options for flexible spot pricing.

Local vs Cloud Cost Hint

Mode	40h / month	120h / month
Local power only (3090 baseline)	$2.24	$6.72
A6000 48GB	$30.4	$91.2

Related Model Profiles

Granite 3.1 MoE 3B Q4 4GB min, 14GB optimal
Granite 3.1 MoE 3B Q5 6GB min, 16GB optimal
Granite 3.1 MoE 3B FP16 16GB min, 28GB optimal

ollama run granite3.1-moe:3b More reasoning models More 4b-class models Benchmark changelog Submit your test result Check local GPU upgrade

We may earn a commission if you click links on this page.

This page currently uses estimated benchmark baselines. Measured data will replace it after validation.