CD45 · validated tick
exploit- Route
- Supervisor allocation mixture
- Observed signal
- Validated state and live SU high-water mark.
- Decision
Continue productive allocation; retain rescue/explore capacity.
Publication snapshot: 21 July 2026. Endpoint tables use the audited six-target TM0.6 benchmark; replay curves cover all six targets and the five archived method arms in the reconciled publication curve table.Not a live scheduler monitor
What the benchmark shows
The primary endpoint is structure-unique strict success (SU) under a common canonical criterion. Sequence diversity and independent confidence scores are shown separately so the dimensions are not conflated.
6benchmark targets
144worker-GPU-h per arm
1,222T-ReX TM0.6 SU
3,799T-ReX unique sequences
1.22×T-ReX / PUCT total SU
Compare methods and targets
Choose a target and endpoint. Counts are evaluated at the nominal 144 worker-GPU-hour endpoint; ipTM-qualified counts are counted over all scored SU representatives.
Target summary
- T-ReX SU
- 301
- PUCT SU
- 283
- Best official arm
- 45
- T-ReX sequence clusters
- 808
- T-ReX SU/GPU-h
- 2.090
Default static view: CD45. Use the controls to filter targets and endpoints.
| Method | Strict successes | Foldseek SU | Unique sequences | ipTM ≥ 0.7 | ipTM ≥ 0.8 | SU/GPU-h |
|---|---|---|---|---|---|---|
| T-ReX | 1,936 | 301 | 808 | 297 | 153 | 2.090 |
| PUCT | 1,851 | 283 | 792 | 280 | 148 | 1.965 |
| Complexa-only | 647 | 45 | 135 | 44 | 20 | 0.312 |
| BindCraft-only | 13 | 2 | 2 | 2 | 1 | 0.014 |
| BoltzGen-only | 22 | 5 | 22 | 5 | 2 | 0.035 |
Campaign replay
Replay reconciled TM0.6 cumulative-SU trajectories by target and method. Curves are downsampled from append-only campaign records; the x-axis is worker-GPU-hours and excludes controller/LLM overhead.
Select a target and method to replay the archived campaign trajectories.
Evidence-linked decision traces
Selected examples show how validated route-state summaries were converted into exploit, rescue or explore allocation. They are representative audit cards—not private chain-of-thought and not an exhaustive dump of every tick.
CbAgo · validated tick
rescue- Route
- Supervisor allocation mixture
- Observed signal
- Validated rescue-rich state near a route-quality boundary.
- Decision
Increase rescue allocation and monitor strict-yield recovery.
SC2RBD · validated tick
explore- Route
- Supervisor allocation mixture
- Observed signal
- Validated productive state with under-sampled route space.
- Decision
Reserve capacity for a novelty probe.
Generated complexes
The default view is a six-target T-ReX nomination panel. Each candidate is a strict-success Foldseek-TM0.6 representative ranked first within the SU top-10 by the balanced joint score (strict margins + AF2 ipTM + AF2 ipSAE). Physics/affinity values are in-silico post-hoc predictions, not experimental measurements.
How to read this page
The endpoint hierarchy mirrors the paper: generation throughput first, independent quality second, candidate nomination last.
Primary endpoint
Strict successpLDDT ≥ 90, iPAE ≤ 7/31 (0.2258), and binder scRMSD < 1.5 Å.
Structure-unique (SU)Strict successes clustered by Foldseek on the binder chain at TM0.6.
Unique sequencesAll strict successes clustered with MMseqs2 at the sequence-identity threshold; this does not replace SU.
Independent scoresAF2 ipTM/ipSAE/actifpTM and other metrics are post-hoc diagnostics, not SU credit.
Fair-comparison rules
Method-only baselinesComplexa, BindCraft and BoltzGen use their official default configurations.
Controller baselinesPUCT and ε-greedy use the same registered action space and execution stack as T-ReX; only proposal/allocation logic differs.
Compute accountingRates use worker GPU-hours. T-ReX LLM/controller GPU time is not added to the worker denominator.
Top-K interpretationTop-10/50/100 panels are selected after strict filtering and structure clustering; threshold counts use the full scored SU pool.
Data provenance
The dashboard keeps endpoint counts, campaign replay and candidate nomination as separate auditable layers. Each layer is rendered only from a reconciled artifact; the page does not claim to be a live scheduler monitor.
| Layer | Scope | Source used here |
|---|---|---|
| Endpoint comparison | Six targets · TM0.6 · 144 worker-GPU-h nominal endpoint | Publication endpoint table |
| Campaign replay | Six targets · T-ReX, PUCT and official-default arms | Reconciled cumulative scaling-curve CSV |
| Candidate panel | Six T-ReX candidates plus two contextual references | Strict-success SU top-10 balanced-rank records and copied PDBs |