reka-edge-2603: benchmark results
Provider: Reka. Access: API.
Aggregate Rating 1234 ± 47, rank #9382 of 10916 rated models, from 63 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| PRISM-VLM - Sycophancy Resistance | 64.1 | Per-axis mean score (%; share of turn-0-correct answers held | 68.3 |
| PRISM-VLM - Hallucination Resistance | 38.6 | Per-axis mean score (%; share of trap questions (absent enti | 26.8 |
| Evals for Every Language - Guarani (gn) | 34.89 | Average Score (%) | 19 |
| PRISM-VLM - Instruction Following | 58.5 | Per-axis mean score (%; mean compliance with 1-3 verifiable | 17.1 |
| Evals for Every Language - Yoruba (yo) | 27.51 | Average Score (%) | 13.6 |
| PRISM-VLM | 27.9 | PScore (%; harmonic mean of the seven per-axis mean scores, | 10 |
| Evals for Every Language - Urdu (ur) | 47.65 | Average Score (%) | 7.7 |
| Evals for Every Language - Macedonian (mk) | 50.26 | Average Score (%) | 7.6 |
| Evals for Every Language - Thai (th) | 45.6 | Average Score (%) | 7.6 |
| PRISM-VLM - Multi-Question Robustness | 8.5 | Per-axis mean score (%; share of items whose 3-5 bundled que | 4.9 |
| Evals for Every Language - Greek (el) | 42.18 | Average Score (%) | 4.5 |
| PRISM-VLM - Counting | 32.9 | Per-axis mean score (%; exact-integer accuracy on synthesize | 3.7 |
Interactive version: theaggregate.ai/model?slug=reka-edge-2603 · How It Works · Data refreshed daily, snapshot 2026-11-08.