IonWarpTry for free AI code review benchmarks
An app has the highest F1 in 5 of the 6 review types where apps were graded; a model leads Docs Review (Gemini 3.8 Flash, 68.6%). Each review type's best model costs $0.01–$0.42 a review in measured inference; the apps' list prices come to about $1.00–$2.00 a review, an estimate on a different basis.
| Review type | Best model | F1 | $ per review | p50 time | Best app | App F1 | PRs | Labels | Snapshot |
|---|---|---|---|---|---|---|---|---|---|
| Code Review | Muse Spark 1.3 | 56.6% | $0.42 | 3m 18s | Qodo v2 | 58.5% | 50 | Independent | |
| Security Review | GLM 5.3 Flash | 66.7% | $0.01 | 1m 38s | Qodo v2 | 73.3% | 13 | Provisional | |
| AI Architecture Review | DeepSeek V4.1 Flash | 47.1% | $0.05 | 6m 14s | CodeRabbit | 72.7% | 11 | Provisional | |
| Log & Errors Review | Gemini 3.8 Flash | 67.7% | $0.10 | 1m 28s | Qodo v2 | 72.7% | 17 | Provisional | |
| Docs Review | Gemini 3.8 Flash | 68.6% | $0.08 | 59 sec | CodeRabbit | 55.6% (11/19 PRs) | 19 | Provisional | |
| API Performance Review | Muse Spark 1.3 | 56.0% | $0.06 | 42 sec | Greptile v4.1 | 76.9% (12/18 PRs) | 18 | Provisional | |
| SOC 2 Review | GLM 5.3 Flash | 66.7% | $0.01 | 34 sec | Not graded | — | 20 | Provisional | |
| CI/CD Performance | DeepSeek V4 Pro | 72.7% | $0.03 | 1m 22s | Not graded | — | 18 | Provisional | |
| Cost Review | DeepSeek V4.1 Flash | 57.1% | $0.07 | 10m 39s | Not graded | — | 20 | Provisional |
A review type's best model is its highest-F1 IonWarp model graded on every PR. Two models that show the same F1 are a tie, and a tie goes to the cheaper one, then the faster one. Review times come from runs on Sep 8–24, 2026 (models). Code Review's PRs and labels come from the independent Martian code-review-bench; every other review type uses provisional labels IonWarp scoped itself. Open a review type for every model and app, ranked by Reviewer Score.