Neuron registration trends
The alpha price in τ, day by day, and what moved it each way.
Log change over the last 30 days −1.18%: protocol buys 0.00%, root sells −0.39%, users net +0.13%, cross-subnet −0.92%, unattributed −0.0003%.
The chart follows the alpha price in τ. Each period starts at its opening price and steps through the mechanisms below, in a fixed order, to its closing price: a step that rises pushed the price up, one that falls pushed it down, and its height is how far it moved the price. A thin line carries each close into the next period’s open; it is dashed where the two differ. The faint candle behind each period is the market’s own open, high, low and close, for context.
With many periods on screen (90 days, or hours), the steps would be too thin to read, so the chart draws the price as a line instead and colours each period by the mechanism that moved it most. Hover a period for the full step-by-step breakdown.
Not here, because they do not move the spot price: the per-block TAO injection into the pool, and burns.
All figures are log changes, not simple percentage changes, so the parts add up exactly to the total; on big moves the two differ (a 10% log fall is a 9.5% price fall).
2026-11-06 UTC, as of ≈04:52 UTC (block 9,444,752)
The day isn’t over yet, so these figures keep changing. They are in no daily total, average or rank.
Production LLM inference requires constant optimization to reduce latency and improve throughput, but evaluating inference servers fairly across different hardware and software stacks is complex and resource-intensive. Teams typically benchmark against fixed baselines in isolation, with no standardized way to compare novel optimizations across a competitive landscape.
A decentralized network of validators running standardized benchmarks against the same baseline on identical hardware creates a fair, transparent leaderboard. Miners submit containerized servers and are evaluated in head-to-head comparison; the fastest correct implementation earns rewards, creating direct economic incentive to improve real-world inference performance.
Cacheon (SN14) implements this as an open competition where miners submit Docker images of inference servers for Qwen2.5-72B-Instruct, validators pull those images and measure end-to-end response time against a vLLM baseline on 8-GPU pods, and the fastest correct server earns 80% of the competition pool. Correctness is verified first; speed improvements that fail greedy-decoding checks receive zero score.
Auto-generated from the GitHub repository and may not reflect the latest changes.
Cacheon (SN14) is a Bittensor subnet running an open competition for production-grade LLM inference optimization. Miners submit containerized inference servers for Qwen2.5-72B-Instruct that are evaluated by validators against a vLLM baseline on 8-GPU pods, with the fastest correct server earning 80% of the competition pool based on end-to-end response time improvement.
Miners build containerized inference servers serving Qwen2.5-72B-Instruct via /v1/chat/completions, commit the image reference and digest on-chain with a fee, and await evaluation. Validators consist of two components: an always-on CPU host that scans the chain and sets weights, and an ephemeral GPU pod rented on-demand that pulls miner images, mounts model weights at /models, and measures end-to-end response time against a vLLM baseline. The fastest correct server becomes the winner and earns the majority of emission.
Score is calculated as speed_improvement = median(max(0, (baseline_e2e - miner_e2e) / baseline_e2e)) if correctness passes, else 0. Challengers must exceed the current winner's fresh score by a fixed 1% margin to prevent noise-driven churn.
The sections above were auto-generated from the GitHub repository and may not reflect the latest changes. View repository (github.com, opens in a new tab, external site)
Inference optimization
This information was auto-extracted from the GitHub repository and may be incomplete or out of date. Always refer to the subnet's GitHub repository (github.com, opens in a new tab, external site) for the latest instructions.
GPU hardware is only needed for local testing; inference server packaged as Docker image (maximum 20 GB) with model weights mounted at runtime
Not yet verified: GPU
Two components: CPU host (always-on, chain scanning, weight setting) and ephemeral GPU pod (eval, rented on-demand when challengers queued)
Machine-extracted from the subnet's repository docs; may be wrong or out of date. Check the repository before buying or renting hardware. Open the repository (github.com, opens in a new tab, external site) · Extracted 28 Sep 2026
Tag your inference server Docker image with your registry path
docker tag my-server:latest docker.io/myuser/cacheon-miner:v1
Push the Docker image to a public container registry
docker push docker.io/myuser/cacheon-miner:v1
Pay submission fee and commit image reference and digest on-chain
python miner/commit.py --wallet-name <wallet> --wallet-hotkey <hotkey> --image "docker.io/myuser/cacheon-miner:v1" --digest "sha256:..." --fee 0.1 --network finney --netuid 14
Clone the Cacheon repository
git clone https://github.com/latent-to/cacheon
Navigate to repository root
cd cacheon
Copy environment template and add wallet and S3 config
cp .env.example .env
Start the always-on CPU host component (chain scanning, weight setting)
docker compose up --build
Provision ephemeral GPU pod on-demand (run when challengers appear)
bash scripts/gpu_setup/setup.sh
Start the GPU pod evaluation component
docker compose -f validator/gpu-compose.yml up --build -d
| Day (UTC) | Share |
|---|---|
| 2026-11-05 | 0.01% |
| 2026-11-04 | 0.01% |
| 2026-11-03 | 0.01% |
| 2026-11-02 | 0.01% |
| 2026-11-01 | 0.01% |
| 2026-10-31 | 0.01% |
| 2026-10-30 | 0.01% |
| 2026-10-29 | 0.01% |
| 2026-10-28 | 0.01% |
| 2026-10-27 | 0.01% |
| 2026-10-26 | 0.01% |
| 2026-10-25 | 0.01% |
| 2026-10-24 | 0.01% |
| 2026-10-23 | 0.01% |
| 2026-10-22 | 0.01% |
| 2026-10-21 | 0.01% |
| 2026-10-20 | 0.01% |
| 2026-10-19 | 0.01% |
| 2026-10-18 | 0.01% |
| 2026-10-17 | 0.01% |
| 2026-10-16 | 0.01% |
| 2026-10-15 | 0.0033% |
| 2026-10-14 | 0.00% |
| 2026-10-13 | 0.00% |
| 2026-10-12 | 0.00% |
| 2026-10-11 | 0.00% |
| 2026-10-10 | 0.00% |
| 2026-10-09 | 0.00% |
| 2026-10-08 | 0.00% |
| 2026-10-07 | 0.00% |
| 2026-10-06 | 0.00% |
| 2026-10-05 | 0.00% |
| 2026-10-04 | 0.00% |
| 2026-10-03 | 0.00% |
| 2026-10-02 | 0.00% |
| 2026-10-01 | 0.00% |
| 2026-09-30 | 0.00% |
| 2026-09-29 | 0.00% |
| 2026-09-28 | 0.00% |
| 2026-09-27 | 0.00% |
| 2026-09-26 | 0.00% |
| 2026-09-25 | 0.00% |
| 2026-09-24 | 0.00% |
| 2026-09-23 | 0.00% |
| 2026-09-22 | 0.00% |
| 2026-09-21 | 0.00% |
| 2026-09-20 | 0.00% |
| 2026-09-19 | 0.00% |
| 2026-09-18 | 0.00% |
| 2026-09-17 | 0.00% |
| 2026-09-16 | 0.00% |
| 2026-09-15 | 0.00% |
| 2026-09-14 | 0.00% |
| 2026-09-13 | 0.00% |
| 2026-09-12 | 0.00% |
| 2026-09-11 | 0.00% |
| 2026-09-10 | 0.00% |
| 2026-09-09 | 0.01% |
| 2026-09-08 | 0.00% |
| 2026-09-07 | 0.00% |
| 2026-09-06 | 0.00% |
| 2026-09-05 | 0.00% |
| 2026-09-04 | 0.00% |
| 2026-09-03 | 0.00% |
| 2026-09-02 | 0.00% |
| 2026-09-01 | 0.00% |
| 2026-08-31 | 0.00% |
| 2026-08-30 | 0.00% |
| 2026-08-29 | 0.00% |
| 2026-08-28 | 0.00% |
| 2026-08-27 | 0.00% |
| 2026-08-26 | 0.00% |
| 2026-08-25 | 0.00% |
| 2026-08-24 | 0.00% |
| 2026-08-23 | 0.00% |
| 2026-08-22 | 0.00% |
| 2026-08-21 | 0.00% |
| 2026-08-20 | 0.00% |
| 2026-08-19 | 0.00% |
| 2026-08-18 | 0.00% |
| 2026-08-17 | 0.00% |
| 2026-08-16 | 0.00% |
| 2026-08-15 | 0.00% |
| 2026-08-14 | 0.00% |
| 2026-08-13 | 0.00% |
| 2026-08-12 | 0.00% |
| 2026-08-11 | 0.00% |
| 2026-08-10 | 0.00% |
| 2026-08-09 | 0.00% |
| 2026-08-08 | 0.00% |
