Verification Transport - ATBench - V-EVIDENCE - Native Operational Balanced Accuracy: leaderboard
Matched native/abstracted/restored executions; 7exactmodelbackbones; protocol andbenchmark split. OBA nativebalancedaccuracy; VTFR conditionedonnativecorrectjudgments; RRR conditionedontransportfailures; RDR restoreddegradation. Denominators differ andmustnotpool. Invalid/noverdict outcomes countedunderpaperoperationalfailurepolicy. Three abstractionboundaries independent. Full immutablepayload pins alldefinitions. Metric: Native rate (%). Source: https://arxiv.org/html/2611.18471v1
Metric: Native rate (%). Source: arxiv.org. Saturation forecast: Around 2030. 7 models tracked. Updated 17 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Flash | 64.5 | #467 |
| 2 | GPT-OSS-120B | 59.8 | #707 |
| 3 | Gemma 4 31B | 57.5 | #385 |
| 4 | Gemini 3.7 Flash | 57.3 | #38 |
| 5 | Llama 4 Maverick | 56.5 | #1118 |
| 6 | GPT-OSS-20B | 55.8 | #1163 |
| 7 | Gemma 4 26B A4B | 55 | #581 |
No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=verification-transport-atbench-v-evidence-native-operational-balanced-accuracy · How It Works · Data refreshed daily, snapshot 2026-11-17.