Athena · SignalRank — public calibration snapshot

Which model should you trust, for which question? Measured, not assumed.

SignalRank runs a known-answer challenge harness across frontier and open models, scores every response against ground truth, and maintains a per-domain trust calibration. This is a static snapshot of the live corpus.

440
known-answer questions
32
calibrated domains
10,791
challenges executed
~19
models under test
The honest number: router accuracy is currently 16.3% (116/713) — the harness measures well; routing on those measurements is the open research problem. Better priors, decay modelling and per-domain sample efficiency are exactly where academic collaboration takes this next. We publish the weak number because the method only works if the numbers are real.
Trust is also an optimisation engine. Look at the market-analysis table below: a model priced at a fraction of the flagships outperforms them in that domain. Evidence-based routing therefore cuts token spend and raises answer quality at the same time — send each question to the model the data says wins it, at the price that model actually costs. At enterprise volume, calibrated routing is a unit-economics instrument, not just a safety one.

Medical

Top of table — clinical reasoning, guidelines, pharmacology
claude-opus-4-7
0.94437
gpt-5-mini
0.93052
claude-opus-4-8
0.92431
grok-4
0.91963

Market analysis

Domain inversion: a small model leads — this is why per-domain calibration exists
gpt-5-mini
0.93810
gpt-5.4
0.92810
grok-4-fast
0.80914
claude-opus-4-7
0.78312

Legal

Statutory interpretation, case analysis, drafting
mimo-v2.5-pro
0.9056
claude-opus-4-7
0.8907
grok-4-fast
0.86410
claude-sonnet-4-6
0.8487

Adversarial robustness

Trick premises, injection attempts, manipulative framing
claude-opus-4-7
0.92262
claude-opus-4-8
0.91617
claude-sonnet-4-6
0.91062
gpt-5.5
0.86517

Temporal / live knowledge

Every model degrades — the harness measures knowledge decay directly
gpt-5.5
0.6813
claude-opus-4-7
0.66513
grok-4
0.59114
deepseek-chat
0.47114

Method, in brief

Every model in the roster answers the same known-answer questions per domain; responses are scored against ground truth by an independent judge model; scores update a per-model, per-domain trust calibration with recency weighting. Human corrections enter the corpus as new ground truth. Low-scoring outliers (visible in the full tables) are retained, not pruned — some reflect integration or refusal-policy artifacts rather than capability, and distinguishing those is itself part of the calibration research.

Roster note: model list reflects the snapshot date; newest-tier additions (including Anthropic's Fable/Mythos class) are onboarded as calibration batches complete.