HolmesAI Pioneer Rewards Season II

HolmesAI Pioneer Rewards Season II will begin on November 1, 2024 (UTC/GMT -4). Season II continues to recruit GPU providers to join the HolmesAI Cloud network and support AI inference with computing power. GPU providers participating in this event will earn points based on their GPU contributions, which will be an important reference for future HolmesAI airdrop allocations.

Limitation: A maximum of 5,000 GPUs will be accepted by the end of Season II. There is no first-come, first-served policy, as we only track the total GPU count. As long as the total hasn’t reached 5,000, participants still have a chance to join.

Duration: November 1, 2024 - December 31, 2024 (UTC/GMT -4)

Participants: GPU computing power providers

Participation Link: https://console.holmesai.xyz/ (opens in a new tab)

Operation Guide: https://holmesai.xyz/docs/user-manual/provider (opens in a new tab)

1. GPU Model Requirements

Participants must own the following GPU model(s) to participate:

NVIDIA L40

NVIDIA L4

NVIDIA L40S

NVIDIA A10

NVIDIA A30

NVIDIA A40

NVIDIA A100

NVIDIA H100

NVIDIA GeForce RTX 3060

NVIDIA GeForce RTX 3070

NVIDIA GeForce RTX 3080

NVIDIA GeForce RTX 3090

NVIDIA GeForce RTX 4060

NVIDIA GeForce RTX 4070

NVIDIA GeForce RTX 4080

NVIDIA GeForce RTX 4090

Tesla P100

Tesla V100

2. Network Requirements

● Download Speed: ≥100Mbps

● Upload Speed: ≥75Mbps

3. Regional Restrictions

GPUs located in the following regions are not supported for now: Mainland China, Russia, Palestine, Cuba, Iran, North Korea, Syria.

4. Points Reward Rules

Points earned through providing computing power are referred to as Miner Points (MP).

Let i denote the round number, where i=1,2,…,N

During a running time T, a miner m with n units of hardware with condition H can earn the following points reward:

MP^i_m=(1-λ)^{i-1}*n*A*H*R*B*T*MPBase*MP Booster

● MPBase: The base points reward mining speed, indicating the number of points that can be mined per hour by running one GPU.

In Season 1, the reward rule is set as

wlMPBase = 10wlMP_m/hour

for consistency:

MPBase=10MP_m/hour

Where:

● λ: Decay function. Providers who join the network earlier will earn more points. For example, points earned in the i-th round will be more than those earned in the i+1-th round. Thus,

MPBase^i = (1-λ)^{i-1}·MPBase

● H: Hardware reward multiplier (coefficients detailed in Appendix 1)

● R: Regional reward multiplier (coefficients detailed in Appendix 2)

● B: Bandwidth reward multiplier (coefficients detailed in Appendix 3)

● T: Effective runtime of the GPU, with a minimum required duration of 1/2 hour (i.e., the GPU must run continuously for at least 1/2 hour to be recorded as one valid run).

● A: Computing power usage reward multiplier. When the computing power provided by a node is effectively used, the mining speed within one hour is multiplied by this factor, where A=160%

● MPBooster: An additional reward multiplier that can increase the points mined per unit time.

Notes

● HolmesAI will not reset the MP earned by computing power providers; providers will retain their earned points even after exiting the network.

● After exiting the network, the time-limited MPBooster for that address will be canceled.

● HolmesAI will review malicious behavior and remove providers that harm the network.

● HolmesAI reserves the right to the final interpretation of the activity.

● If you have any questions or need assistance, please email: support@holmesai.xyz.

Join our Telegram community: https://t.co/naNXmnvlw5 (opens in a new tab)

Join our Discord community: https://t.co/GKQULhRSVc (opens in a new tab)

Follow HolmesAI: https://x.com/HolmesAI_ (opens in a new tab)

Appendix 1: GPU Model Coefficients

Appendix 2: Regional Coefficients

Appendix 3: Bandwidth Coefficients