Ironbee Raw Code Review - Frontier - F1: leaderboard
Ironbee raw code review (2026): standalone model review of real pull requests, three repeated runs. Full uses 50 PRs and Frontier 10 PRs; no cohort pooling. Model-specific requested tiers remain explicit; precision, recall, F1 and F2 are separate metrics. Aggregated source fractions are multiplied by 100, with full precision retained. F1 (%).
Metric: F1 (%). Source: raw.githubusercontent.com. Saturation forecast: Rough model projection: around 2026. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Ironbee Study claude-fable-5.1 - low tier | 59.17 | |
| 2 | Ironbee Study claude-fable-5.1 - default tier | 51.21 | |
| 3 | Ironbee Study gpt-6-astra - low tier | 50 | |
| 4 | Ironbee Study claude-fable-5.1 - xhigh tier | 49.88 | |
| 5 | Ironbee Study gpt-6-astra - default tier | 45.22 | |
| 6 | Ironbee Study gpt-6-astra - xhigh tier | 43.31 |
Interactive version: theaggregate.ai/benchmark?slug=ironbee-raw-code-review-frontier-f1 · How It Works · Data refreshed daily, snapshot 2026-10-21.