freeintelligence.ai · tool

GPU Pricing Comparison

Live per-hour pricing across Vast.ai, Salad, Terra Compute, and Hashrate — from RTX 3060 to B200. Updated July 14, 2026.

Sources: Vast.ai, Salad, Terra Compute, Hashrate, computeprices.com, gpuperhour.com

Per-Hour Pricing Matrix

GPUVRAMVast.aiSaladTerra ComputeHashrate
RTX 509032GB GDDR7$0.42–$0.48$0.25Listed
RTX 409024GB GDDR6X$0.35–$0.39$0.11–$0.18Listed
RTX 3090 Ti24GB GDDR6X$0.25–$0.35$0.10–$0.12Listed
RTX 309024GB GDDR6X$0.20–$0.30$0.09–$0.10Listed
RTX 308010GB GDDR6X~$0.12$0.06Listed
RTX 30708GB GDDR6~$0.08$0.05Listed
RTX 306012GB GDDR6~$0.08$0.05–$0.08Listed
RTX 4070 Ti Super16GB GDDR6X~$0.15$0.07–$0.10Listed
RTX 5070 Ti16GB GDDR7~$0.18$0.07–$0.17Listed
RTX 408016GB GDDR6X$0.20–$0.28$0.08–$0.12Listed
RTX 4060 Ti 16GB16GB GDDR6~$0.10$0.06–$0.08Listed
H100 SXM80GB HBM3$0.90–$2.27Multi-GPUListed
H200141GB HBM3e$2.15–$6.00Multi-GPUListed
B200$4.75–$5.08Listed
B300$5.85–$6.30Listed
A100 80GB80GB HBM2e$0.75–$4.00Multi-GPUListed
V100 32GB32GB HBM2~$0.50–$1.00Listed
Tesla P4024GB GDDR5~$0.15–$0.25Listed

Best price highlighted in green. Hashrate is a comparison index, not a provider. Terra Compute sells bare-metal server hosting at $0.075/kWh. Prices fluctuate — check each provider for current rates.

Provider Profiles

🥬 Salad

$0.05–$0.25/hr
60,000+ daily active GPUs. Distributed network (gamers' machines). Cheapest for consumer GPUs. Best for batch inference on RTX 3090/4090/5090 class cards.

🌐 Vast.ai

$0.08–$6.30/hr
20,000+ GPUs, RTX 3060 → B300. Marketplace model — prices vary by host. Widest selection. Best for datacenter GPUs (H100, B200, A100).

🏗️ Terra Compute

$0.075/kWh
Bare-metal server hosting. You buy the server ($4k–$8k), they provide power/cooling/colo. Best for sustained high-utilization multi-GPU workloads.

Rent vs. Buy — Break-Even Analysis

HardwareUpfrontRental EquivalentBreak-Even
Used RTX 3090 24GB$700$0.20–$0.30/hr on Vast.ai2–4 months
Dual RTX 3090 48GB$1,700$0.40–$0.60/hr3–5 months
Mac Studio M3 Ultra 96GB$4,199$0.50–$0.90/hr (H100)8–14 months
Mac Studio M3 Ultra 192GB$7,499$1.50–$2.27/hr (H100)5–8 months
4× Tesla P40 96GB~$1,200$0.60–$1.00/hr2–3 months

Assumes 24/7 operation. Break-even = months until owning costs less than renting the equivalent GPU. After break-even, you pay only electricity ($12–$50/mo depending on hardware).

What Each GPU Runs

VRAMExample GPUsMax Model Size (Q4)Example Models
8–12 GBRTX 3060, 3070, 4060 Ti8BLlama 3 8B, Mistral 7B, Gemma 7B
16 GBRTX 4070 Ti, 4080, 5070 Ti13BCodeLlama 13B, Mistral 13B, Qwen 14B
24 GBRTX 3090, 4090, 5090, P4027B Q4 / 70B Q3Qwen 3 27B, GPT-OSS 20B, Llama 3.3 70B Q3
48 GBDual 3090, A600070B Q4Llama 3.3 70B, DeepSeek V3 Q4, Qwen 3 72B
80–96 GBH100, A100, Mac Ultra 96GB120B Q4DeepSeek V3, Llama 4, Nemotron
192 GBMac Ultra 192GB, 4× P40200B+ Q3Any model that fits in memory
💡 Recommendation

For 24/7 inference on 27B-class models, buy a used RTX 3090 ($700) — it pays for itself in 2–4 months vs. renting. For variable workloads, rent on Salad (cheapest consumer GPUs). For 70B+ models, rent H100 on Vast.ai or buy dual 3090s ($1,700). For silent 24/7 operation, Mac Studio M3 Ultra 96GB ($4,199).

Related