Benchmark models.
Reference implementations the tournament publishes. Climb past them to earn alpha. Open-source code, deterministic outputs, weekly updates.
| Model | Arch | Target | Mean CORR | Mean AIMC | Avg payout | Source |
|---|---|---|---|---|---|---|
lgb_cho_oyu | LGB | target_everest_20 | 0.0459 | 0.0428 | +0.0190 | |
lgb_dart | LGB | target_everest_20 | 0.0626 | 0.0358 | +0.0175 | |
v0_sherpa_8 | V0 | target_everest_20 | -0.0114 | 0.0160 | +0.0167 | |
v0_sherpa_7 | V0 | target_everest_20 | -0.0103 | 0.0133 | +0.0167 | |
mlp_primary | MLP | target_everest_20 | 0.0608 | 0.0237 | +0.0153 | |
ridge_primary | RIDGE | target_everest_20 | 0.0677 | 0.0360 | +0.0136 | |
elasticnet_primary | ELASTICNET | target_everest_20 | 0.1166 | 0.1057 | +0.0125 | |
xgb_3target | XGB | target_everest_20 | 0.0261 | 0.0168 | +0.0052 | |
lgb_3target | LGB | target_everest_20 | -0.0071 | -0.0154 | +0.0017 | |
lgb_multitask | LGB | target_everest_20 | -0.0007 | -0.0188 | -0.0027 | |
rf_primary | RF | target_everest_20 | 0.0194 | -0.0375 | -0.0125 | |
lgb_kangchen | LGB | target_everest_20 | 0.0190 | -0.0144 | -0.0125 | |
xgb_primary | XGB | target_everest_20 | 0.0079 | -0.0301 | -0.0125 | |
v0_sherpa_3 | V0 | target_everest_20 | -0.0481 | -0.0245 | -0.0167 | |
v0_sherpa_9 | V0 | target_everest_20 | -0.0645 | -0.0511 | -0.0167 | |
v0_sherpa_5 | V0 | target_everest_20 | -0.0205 | -0.0068 | -0.0167 | |
v0_sherpa_10 | V0 | target_everest_20 | -0.0350 | -0.0342 | -0.0167 | |
v0_sherpa_2 | V0 | target_everest_20 | -0.0203 | -0.0018 | -0.0167 | |
v0_sherpa_12 | V0 | target_everest_20 | -0.0682 | -0.0707 | -0.0167 | |
v0_sherpa_1 | V0 | target_everest_20 | -0.0297 | -0.0124 | -0.0167 | |
v0_sherpa_11 | V0 | target_everest_20 | -0.0590 | -0.0550 | -0.0167 | |
v0_sherpa_4 | V0 | target_everest_20 | -0.0995 | -0.1192 | -0.0167 | |
v0_sherpa_6 | V0 | target_everest_20 | -0.0539 | -0.0560 | -0.0167 | |
lgb_primary | LGB | target_everest_20 | -0.0892 | -0.0867 | -0.0273 |
Round-by-round α
Benchmark time-series will appear as rounds resolve.
Benchmark models produce fully deterministic outputs. Given the same live feature set, every run returns identical predictions. No random seeds, no cloud state, no stale caches. You can reproduce any round's benchmark score from the public repository with a single command.
All benchmark code is published on GitHub under the Everesteer organisation before each season opens. The repository includes training scripts, frozen weight files, and a validation harness that matches the tournament's internal scorer.
"Beating" a benchmark means your model's tournament-settled alpha exceeds the benchmark's alpha in the same round, net of any correlation penalty. The tournament tracks a 20-round rolling beat-rate. Models above 60% against eiq_minera_model qualify for the Top 14 consideration pool.
Benchmarks update weekly on the Wednesday before each round opens. The update pulls the latest live feature set, re-runs inference, and commits new predictions to the repository within two hours. Benchmark scores appear in the leaderboard alongside participant models.
The eiq_minera_model benchmark uses stake as weight: each of the top 14 models' predictions contributes proportionally to its current stake balance. This makes it a live measure of collective conviction rather than a static average.
To propose a new benchmark for inclusion, open a pull request to everestquant/benchmarks with a deterministic inference script, a frozen weight artifact under 50 MB, and a backtest report covering at least 100 rounds.