Loading
Loading

Agentic RL as a Service, Optimize agent trajectories to make agents cheaper, safer, and more reliable.
Indicators are computed from daily candles and refresh once per day — directional context, not trading advice.
Recent emission events for this subnet
| Block | UID | Type | TAO | Alpha |
|---|---|---|---|---|
| Time↓ | Type | Event | τ Amount | λ Amount | USD Value | Account |
|---|---|---|---|---|---|---|
| Sell | 9117110-208 | 1.2108 τ | 147.5351λ | — | ||
| Sell | 9117070-247 | 0.8441 τ | 102.8404λ | — | ||
| Sell | 9117023-294 | 3.5908 τ | 437.4369λ | — | ||
| Sell | 9116858-363 | 0.0785 τ | 9.5639λ | — | ||
| Buy | 9116540-152 | 0.4300 τ | 52.369λ | — | ||
| Sell | 9116415-200 | 2.4228 τ | 295.0874λ | — | ||
| Sell | 9116308-203 | 27.715 τ | 3,374.85λ | — | ||
| Sell | 9116267-281 | 0.2163 τ | 26.2722λ | — | ||
| Sell | 9116259-187 | 0.6079 τ | 73.8294λ | — | ||
| Buy | 9115939-243 | 0.2000 τ | 24.2886λ | — | ||
| Buy | 9115914-191 | 3.4708 τ | 421.5144λ | — | ||
| Sell | 9115824-207 | 17.9506 τ | 2,180.78λ | — | ||
| Sell | 9115671-177 | 4.8667 τ | 590.2439λ | — | ||
| Sell | 9115286-169 | 0.0025 τ | 0.3066λ | — | ||
| Sell | 9115265-158 | 0.0135 τ | 1.6418λ | — |
This information was auto-extracted from the GitHub repository and may be incomplete or out of date.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 for the latest instructions.
Miners do not run the evaluation harness. They write SKILL.md and optionally a fusion policy (policy.py or policy.json), build a pack, and submit via web API. Validators run all evaluation on their infrastructure. Local testing via scripts/eval_pack.py requires Docker and an Engy API key.
Validators run the trajrl-bench harness (scenario containers + verifier) in Docker. No GPU required for the agent—Hermes is CPU-based. LLM inference (judge, evals) is metered through an external API (Engy). ~10 GB disk for scenario images on first run. One eval per challenger epoch (~20–30 min) costs ~$0.5–10 in LLM tokens depending on policy complexity.
Create a Bittensor wallet (one-time)
pip install bittensor-cli && btcli wallet create --wallet.name miner --wallet.hotkey default
Register on subnet 11 (dynamic cost; check balance first)
btcli subnet register --netuid 11 --wallet.name miner --wallet.hotkey default
Clone and install trajectoryrl package
git clone https://github.com/trajectoryRL/trajectoryRL && cd trajectoryRL && pip install -e .
Copy miner environment template (optional but recommended for local testing)(optional)
cp .env.miner.example .env.miner
Build a pack from SKILL.md (SKILL.md-only submission uses default policy)
trajectoryrl-miner build SKILL.md -o pack.json
Build a pack with a fusion policy directory (Season 2; policy.py or policy.json + helpers)(optional)
trajectoryrl-miner build SKILL.md --policy ./my_policy -o pack.json
Validate pack schema and size (≤32 KB) locally before submission
trajectoryrl-miner validate pack.json
Test with the exact validator harness (Docker + your Engy key; local runs cost from your account)(optional)
LLM_API_KEY=$ENGY_API_KEY python scripts/eval_pack.py --pack pack.json -o ./eval_output
Submit pack to the platform (requires 50 α recycle fee when enabled; signed with hotkey)
trajectoryrl-miner web-submit pack.json
Check on-chain commitment or latest submission status(optional)
trajectoryrl-miner status
Create a Bittensor wallet (one-time)
pip install bittensor-cli && btcli wallet create --wallet.name validator --wallet.hotkey default
Register on subnet 11 as a validator
btcli subnet register --netuid 11 --wallet.name validator --wallet.hotkey default
Stake at least 100 TAO to become eligible as a validator
btcli stake add --wallet.name validator --hotkey default --netuid 11 --amount 100
Clone and install trajectoryrl package
git clone https://github.com/trajectoryRL/trajectoryRL && cd trajectoryRL && pip install -e .
Verify Docker daemon is installed and running
docker --version
Copy validator environment template and set WALLET_NAME and LLM_API_KEY (Engy key)
cp .env.validator.example .env.validator && vim .env.validator
Start validator daemon (includes auto-update via Watchtower)
docker compose -f docker/docker-compose.validator.yml --env-file .env.validator up -d
Monitor validator logs(optional)
docker compose -f docker/docker-compose.validator.yml logs -f validator
| Fund | Alpha held | Realizable | Share of fund | Weight in vector |
|---|---|---|---|---|
| Arbos5GTWwu…ySXP | 2.5Kλ | 20.39τspot 20.41τ | 3.27% | 6.25% |
| tao.bot5E2LP6…eZ5u |
| 1.7Kλ |
| 13.73τspot 13.74τ |
| 0.23% |
| — |
| Taostats5GKH9F…TrP1 | 1.0Kλ | 8.50τspot 8.51τ | 0.25% | — |
| Yuma, a DCG Company5DXdHi…L1S1 | 844.06λ | 6.93τspot 6.93τ | 0.32% | — |
| Crucible Labs5CAQAg…7FmJ | 812.53λ | 6.67τspot 6.67τ | 0.50% | — |
| 5Dd8ga…GrWv | 787.19λ | 6.46τspot 6.46τ | 0.45% | — |
| General Tensor5Ggasr…xGTV | 720.24λ | 5.91τspot 5.91τ | 0.64% | 2.00% |
| Kraken5Ckaof…uUes | 681.19λ | 5.59τspot 5.59τ | 0.57% | — |
| Openτensor Foundaτion5G3wMP…5KP3 | 582.10λ | 4.78τspot 4.78τ | 0.44% | — |
| tao55CsvRJ…2zVp | 545.31λ | 4.48τspot 4.48τ | 0.41% | — |