Know what you pay before you ship.
Published catalog rates, adjusted for demand and capacity. Providers earn a transparent share of each completed job.
- Per-token billing
- Catalog rates published
- Live dashboard
Developers
Per-token inference.
Billed per million input and output tokens. Rates vary by model and demand. Compare usage in Scalattice Cloud.
# Per-million tokens (live catalog)
qwen-2.5-coder-7b in $0.029 out $0.077
deepseek-r1-7b in $0.040 out $0.100
gemma-3-27b in $0.065 out $0.129
qwen-3-32b in $0.066 out $0.242
qwen-3-1.7b in $0.083 out $0.083
qwen-3-8b in $0.086 out $0.173
qwen-3-14b in $0.091 out $0.192
llama-3.3-70b in $0.098 out $0.306
# Full catalog & live rates → Scalattice Cloud
Qwen
Llama
DeepSeek
Gemma
Providers
Earn per job served.
You set availability. Scalattice routes paying inference to your hardware. Payouts tracked in the dashboard with no connection fees.
Revenue split
The majority of developer token spend on each completed inference job.
On demand
Request a payout any time your available balance is above the minimum threshold.
Per-machine schedule
Set availability windows for each GPU from Scalattice Cloud.
# Provider dashboard (example)
jobs completed 847
tokens served 12.4M
earnings $284.50
available $42.50
# request payout any time (min $10)Enterprise
Volume and committed use.
Contact us for committed capacity, custom regions, and invoicing. Enterprise features are arranged with our team.
Committed capacity
Reserve GPU pools for predictable latency and spend.
Custom regions
Deploy to jurisdictions your compliance team requires.
Invoicing
Annual contracts and PO-based billing for finance teams.