What Dolphin Is
Dolphin ($POD) aggregates distributed GPU supply and routes model inference work to nodes, turning idle compute into serving capacity for open-source model demand.
Dolphin
Inference Economics
Dolphin aggregates GPU supply and routes inference work to nodes; all while maintaining quality, unlocking demand and compounding network growth.
Cost per 1M output tokens
Overview
Why OpenRouter Matters
OpenRouter can route users toward the cheapest reliable provider. If Dolphin sits below the market floor while keeping quality intact, latent demand can convert into paid token volume.
Pricing Note
The model uses a $1.00 / 1M output-token floor and $0.15 / 1M input-token pricing, then shows how discounted Dolphin scenarios affect revenue and margin.
Reference Scope
Qwen 35B is an illustrative workload, not the full Dolphin model surface. Dolphin can serve a broader model mix, so the numbers are directional economics rather than an exhaustive forecast.
Interactive Model
Inspect how tokens, pricing, emissions, and GPU capacity move together.
Change the output-token demand, input/output ratio, and pricing scenario to see daily revenue, emissions burden, economic margin, and implied GPU requirements.
Total notional revenue per day
Click a point or change the controls to update the selected economics.
Floor$1.00 out / $0.15 in
Price
Output
Ratio
Total daily revenue
$3.50K output + $10.50K input
$POD emissions
$0.50 per 1M output tokens
Revenue after emissions
87.5% economic margin
Implied GPUs
300 GPU observed benchmark
Revenue bridge87.5% margin
Total revenue$14.00K
$POD emissions$1.75K
After emissions$12.25K