Lab
138 explorable, interactive pieces — every concept we write about, made tangible.
Cost per million tokens and throughput for Llama 3.3 70B on 8x H100 as the batch fills. A single-tenant node (B=1) runs ~25 tok/s at ~$258/M; a filled batch (B=256) runs ~2,800 tok/s at ~$2.30/M — same silicon, 112x cheaper. Pick a batch; read the tax you pay for the GPU you can't fill.
In decentralized RL the rollout worker acts with a policy several steps behind the trainer. Staleness g = broadcast time / step cadence. Pick a model and link, toggle sparse deltas, and watch g cross INTELLECT-2's demonstrated 4-step budget — full 32B weights blow right past it.
Bittensor's Yuma Consensus scores a miner by the stake-weighted median of validator weights, then clips everything above it. Validators sit on a cumulative-stake axis; the consensus line is read at the κ=50% mark. Push a colluding bloc past 50% and watch the pump go from clipped to unstoppable.
An inspection game between a solver and a verifier. Drag the operating rate and tune the bond, reward, and bounty to find p*, the minimum sampling rate that makes honest AI inference a dominant strategy — and watch it collapse without a bounty.
An on-chain agent is told to pay an invoice, but one MCP tool hides an instruction in its description that redirects the transfer to an attacker. Pick a defense layer — model, tool, client, policy, custody — and run it; alignment alone refused under 3% of these in MCPTox.
Cumulative net carry of a $10k delta-neutral perpetual funding harvest over real Hyperliquid funding regimes. The line starts below zero by the round-trip fee, then climbs at the annualized funding rate; a flip slider rolls it over. Carry, not alpha — until funding inverts.
On-chain cost of a query versus how much history it touches. Re-executing pays a cold SLOAD per value and hits the 60M block-gas wall — or can't reach the data at all. A coprocessor's proof verifies flat at ~300k gas whatever the size. Drag the cursor or pick a real query; the cyan line never moves.
Seven ways to verify what an AI agent claims it did, plotted by latency added against the share of fabricated claims caught. Pick a claim type and watch it route to the cheapest sufficient verifier — a free chain receipt, a 12ms signed receipt, or a zkLLM proof reserved for the inference itself.
A perpetual market maker's quote ladder. Avellaneda-Stoikov skews only for inventory; funding adds a second, signed skew that survives even at zero inventory. Drag the reservation line to set inventory, push funding and horizon, and watch the two skews stack or cancel.
Decode speed of a frontier MoE model versus interconnect bandwidth. Each layer's expert routing is an all-to-all shuffle bounded by the slowest link. Drag from NVLink down to home broadband: the line crosses the interactive floor at the datacenter edge, then falls to seconds per token.
The on-chain gas cost of verifying a TEE attestation quote, across three paths: naive Solidity (~4.5M), the EIP-7951 precompile (~3.5M), and a zkVM-compressed proof (flat ~290k). Tap a bar and scale to attestations-per-day — the flat proof is the only path that survives continuous agent attestation.
Train a model on its own output, generation after generation. Each step draws a finite sample of size M and refits — so the tails starve first and the curve collapses toward a spike. Toggle replace vs accumulate: replace drives σ̂ toward zero, accumulate holds it. A copy of a copy of a copy.