Synthwave Community Cluster — Llama 3 70B

Consumer RTX 3090 collective — best rates for open models.

Hardware Specifications

GPU Model

NVIDIA RTX 3090

Count

12x

VRAM

288 GB

Uptime

97.2%

Active Model

The inference model currently loaded into VRAM on this node.

Meta · llama-3-70b

8,192 tokens context

Meta's open-weight 70B parameter model, self-hostable.

28,400

Jobs Completed

71,600k

Tokens Processed

Uptime & Reliability

Derived from job completion history. Updated in real time.

Uptime

97.2%

Avg Latency

~140ms

Total Jobs

28,400

Member Since

Jul 2026

Pricing

Choose your instance type. No minimums.

$0.1008/ 1k tokens

Always available. Pay as you go with no commitment — jobs are never interrupted. Ideal for latency-sensitive workloads.

Provider Rate$0.0008

Platform Fee$0.1000

Total$0.1008

Billed at job completion

SLA backed by provider stake

Requires an active buyer account with credits.

Submit a Job

Send a prompt directly to this provider. No API key required — sign in to get started.