Runyard is a free hardware-aware AI model browser. You enter your CPU, GPU, and VRAM and it instantly shows every local LLM that will run on your machine, ranked by speed and quality.

How much VRAM do I need to run local LLMs?

8GB of VRAM runs 7B models like Llama 3.1 8B and Mistral 7B at Q4 quantization. 16GB unlocks 13B models. 24GB lets you run Mixtral 8x7B and Llama 3 70B at lower quantization.

What is the best local LLM for my GPU?

Use Runyard at www.runyard.dev — enter your GPU and VRAM and the Model Radar will rank every compatible LLM for your exact hardware, showing estimated tokens per second for each model.

Can I run Llama 3 locally?

Yes. Llama 3.1 8B at Q4 runs on any 8GB VRAM GPU. Llama 3.1 70B needs around 40GB VRAM at Q4, or an Apple Silicon Mac with 64GB+ unified memory.

AMD MI325X Rental Price — $3.80/hr median | GPU Index

AMD MI325X pricing by provider

One card per provider, cheapest first. A configuration is one priced unit — an instance type, a SKU, a marketplace offer — and the per-GPU price is that unit's price divided by its GPU count. Open a card to see every configuration with its vCPU, RAM and region where the provider publishes them.

In stock only1 providers · 1 configurations

DigitalOcean GPU DropletsIn stock

USA1 config1xcollected 2026-10-11

On-Demand from $3.80

From

$3.80 / GPU / hr

On-DemandVisit website →

What fits on a AMD MI325X

Weights + KV cache at 8K context + runtime overhead, against 230 GB usable of the 256 GB on the card. Capacity follows total parameters, even for mixture-of-experts models — any expert may be needed next. Where one card is not enough, the table says how many are, and prices the set at today's median.

Model	Params	4-bit	8-bit	16-bit
Kimi K3	2.8T104B active	7× GPU$26.60/hr	13× GPU$49.40/hr	25× GPU$95.00/hr
DeepSeek V4.1-Flash	552B16B active	2× GPU$7.60/hr	3× GPU$11.40/hr	5× GPU$19.00/hr
Llama 4 Maverick	400B17B active	✓ 228 GB~502 tok/s · $2.1/M tok	2× GPU$7.60/hr	4× GPU$15.20/hr
gpt-oss-120b	117B5.1B active	✓ 67.9 GB~1673 tok/s · $0.63/M tok	✓ 126.4 GB~886 tok/s · $1.19/M tok	2× GPU$7.60/hr
Llama 3.1 70B	70B	✓ 44.5 GB~122 tok/s · $8.65/M tok	✓ 79.5 GB~65 tok/s · $16.24/M tok	✓ 145.1 GB~34 tok/s · $31.05/M tok
Gemma 4 31B	31B	✓ 21.2 GB~275 tok/s · $3.84/M tok	✓ 36.7 GB~146 tok/s · $7.23/M tok	✓ 65.7 GB~77 tok/s · $13.71/M tok
Qwen3 Coder 30B-A3B	30.5B3.3B active	✓ 19 GB~2586 tok/s · $0.41/M tok	✓ 34.3 GB~1369 tok/s · $0.77/M tok	✓ 62.9 GB~727 tok/s · $1.45/M tok
Qwen3.8 27B	27B	✓ 18.7 GB~316 tok/s · $3.34/M tok	✓ 32.2 GB~167 tok/s · $6.32/M tok	✓ 57.6 GB~89 tok/s · $11.86/M tok
gpt-oss-20b	21B3.6B active	✓ 13.7 GB~2370 tok/s · $0.45/M tok	✓ 24.2 GB~1255 tok/s · $0.84/M tok	✓ 43.9 GB~667 tok/s · $1.58/M tok
Llama 3.1 8B	8B	✓ 6.9 GB~1067 tok/s · $0.99/M tok	✓ 10.9 GB~565 tok/s · $1.87/M tok	✓ 18.4 GB~300 tok/s · $3.52/M tok

tok/s is decode throughput from memory bandwidth, derated 20%, single stream — an estimate, not a benchmark. $/M tok is the hourly median divided by that rate: the one-user floor for a million generated tokens. Set cost uses today's median × GPUs required.

Specifications

Memory256 GB HBM3e

Memory bandwidth6,000 GB/s

ArchitectureCDNA 3

VendorAMD

ClassDatacenter

Released2024-10

Bandwidth is listed because it governs generation speed: each token streams the active weights once, so tokens per second is bandwidth divided by bytes read. TFLOPs decide prompt processing and training, not how fast text appears.

Questions

How much does it cost to rent a AMD MI325X?

As of 2026-10-11, the median on-demand price across 1 providers is $3.80 per GPU-hour, about $2,736 a month at 720 hours. The cheapest we saw was $3.80 at DigitalOcean GPU Droplets.

Why do prices for the same GPU vary so much?

Hyperscalers bundle CPU, RAM and network into an instance price and charge for the ecosystem around it. Neoclouds sell the GPU more directly. Marketplaces are independent hosts competing on price, with the trade-offs of shared, third-party hardware. The per-GPU figure is the instance price divided by GPU count, which is the only way to compare an 8-GPU node with a single-card listing.

Is the median the price I will pay?

No — it is the middle of the market. Each provider's own median is taken first so a provider with many instance sizes counts once, then the median across providers. Spot and reserved rates are excluded from it; they set the floor shown separately. Use it to judge whether a quote is high or low, then verify with the provider.

Which models can a AMD MI325X run?

With 256 GB per GPU, the table above shows which reference models fit on one card at 4, 8 and 16-bit, using the same memory arithmetic as the rest of Runyard: weights plus KV cache plus runtime overhead, with capacity following total parameters even for mixture-of-experts models. Where a model needs more than one GPU it says how many and what that set costs at today's median.

Own the hardware instead? Work out what your card holds with the same arithmetic.

Check what fits →

AMD MI325X

At a glance

AMD MI325X pricing by provider

What fits on a AMD MI325X

Buy or rent a MI325X?

Specifications

Alternatives and comparisons

Questions