JT-35B-Flash: benchmark results
China Mobile's flash-tier 35B dense model from the Jiutian family, positioned for cost-efficient high-throughput inference. Provider: China Mobile. Released 2026-05-14. Access: API.
Unified ELO 1617 ± 1, rank #365 of 1761 rated models, from 18 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA TAU-2 Bench | 99.12 | Accuracy (%) | 99.9 |
| AA GPQA Diamond | 82.93 | Accuracy (%) | 72.4 |
| AA Terminal-Bench Hard | 28.79 | Accuracy (%) | 70.8 |
| Artificial Analysis Intelligence Index | 21.79 | Intelligence Index | 68 |
| BenchmarkList ECI | 125.21 | Capability Index (ECI) | 67.8 |
| AA Omniscience - Law | 17.99 | Accuracy (%) | 67.4 |
| AA Omniscience - Humanities & Social Sciences | 26.01 | Accuracy (%) | 67.3 |
| AA Omniscience - Software Engineering (SWE) | 35.12 | Accuracy (%) | 66.7 |
| AA Long Context Reasoning | 65.67 | Accuracy (%) | 61.8 |
| AA Omniscience | -24.22 | Score | 60.6 |
| AA Omniscience - Business | 19.71 | Accuracy (%) | 60.6 |
| AA-Omniscience Accuracy | 24.18 | Accuracy (%) | 60.4 |
Interactive version: theaggregate.ai/model?slug=jt-35b-flash · How It Works · Data refreshed daily, snapshot 2026-09-05.