Reward Model
Rewarder V3 top-of-book eligibility, fixed market budgets, quantity scaling, spread scaling, and removal.
Rewarder V3 pays only the canonical best bid and best ask. Each side has its own immutable budget and activation time, while reward size depends on elapsed time, live sold quantity, and the distance between the two best prices.
Eligible position
An order accrues only while it is the canonical top of its side. Price priority is absolute: an inferior order does not earn even when it offers more size.
When placement, fill, or cancellation changes the top, the Router reports the transition to Rewarder V3. The rewarder settles the outgoing interval, stores the new cursor, and checkpoints the opposite side when the spread changes.
Fixed market allocation
| Pool | Bid maker sells | Ask maker sells | Per-side cap | Total funding | Duration |
|---|---|---|---|---|---|
| NVDA/USDG | USDG | NVDA | 50,000,000 2DEEP | 100,000,000 2DEEP | 365 days |
The complete allocation is minted into the rewarder at market deployment. Claims transfer from this prefunded balance; the rewarder is not a token minter.
Reward calculation
For each eligible interval, Rewarder V3:
- computes the side's logarithmic scheduled ceiling;
- applies a quantity factor from zero to one;
- applies the current spread multiplier from zero to one;
- credits the verified top-order claimant; and
- caps lifetime accrual at 50 million 2DEEP per side.
An order at or above the target size receives the full quantity factor. A zero-tick spread receives the full spread factor. Additional size cannot exceed the scheduled ceiling.
Spread multiplier
| Best bid/ask spread | Reward multiplier |
|---|---|
| 0 ticks | 100% |
| 10,000 ticks | 10% |
| Maximum representable spread | 0% |
The final 10,000 ticks use a steep inverse-log-style ramp from 10% to 100%. Beyond that region, the multiplier declines gradually toward zero. Both pool sides must be hook-active so any change to either top price can checkpoint the other side under the previous spread.
Quantity targets
The Factory accepts whole-token maximums, reads each token's decimals, and converts those values into raw units. Each side starts at one whole sold token and grows geometrically to its configured maximum over 30 days. A maximum may not exceed 1,000,000× the one-token start.
For NVDA/USDG, the maximums are 1,000,000 USDG and 5,000 NVDA.
Replacement and removal
Deploying a replacement V3 rewarder on an active pool installs the new hook and freezes the prior rewarder's accrual. Its already-accrued claims remain payable from its balance.
Permanent removal is different: the Factory unlinks the hook, freezes both cursors, burns every unearned 2DEEP token in the rewarder, and leaves exactly the accrued-but-unpaid claim liability. The Factory records the pool as permanently removed and will not deploy it again.
See Emission Schedule, Full-Reward Quantity, and Claiming Rewards.