Pokee-Isaac 28B Report - RULER 512K: leaderboard
RULER synthetic long-context retrieval, multi-hop tracing and aggregation tasks at 512K tokens, scored with NVIDIA's data pipeline over all 13 task configurations, 10 samples each. Only models that completed this length are listed; NVIDIA's self-reported Nemotron values are left out. Score (%). From Pokee AI's Pokee-Isaac 28B technical report (September 10, 2026 revision): Pokee ran every listed model itself through hosted APIs under one harness, with each peer's default temperature (unless a benchmark section fixes one) and a high thinking level where available; exhausted retries, context overflows and verifier failures count as zero. Higher is better.
Metric: Score (%). Source: pokee.ai. Saturation forecast: Around February 2027. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.5 Flash Lite | 94.6 | #300 |
| 2 | GPT-5.6 Luna | 91.4 | #117 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=pokee-isaac-28b-report-ruler-512k · How It Works · Data refreshed daily, snapshot 2026-11-12.