Recent Papers / arXiv:2602.10604
Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters
Authors pending
Tasks
editResults
19 results reproduced from this paper.
Sorted instantly in-page
Results
17
SOTA rows
1
Models
3
Datasets
0
| # | Model | Vendor | Benchmark | Value | SOTA | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | Step-3.5-Flash PaCoRe | — | AIME 2025 | 99.9% | #1 | — | source ↗ |
| 02 | Step-3.5-Flash | — | AIME 2025 | 97.3% | — | source ↗ | |
| 03 | Step-3.5-Flash Base | — | HellaSwag | 90.2% | — | source ↗ | |
| 04 | Step-3.5-Flash Base | — | BIG-Bench Hard | 88.2% | — | source ↗ | |
| 05 | Step-3.5-Flash Base | — | GSM8K | 88.2% | — | source ↗ | |
| 06 | Step-3.5-Flash | — | Tau2-Bench | 88.2% | — | source ↗ | |
| 07 | Step-3.5-Flash Base | — | MMLU | 85.8% | — | source ↗ | |
| 08 | Step-3.5-Flash PaCoRe | — | GPQA Diamond | 85.0% | — | source ↗ | |
| 09 | Step-3.5-Flash | — | GPQA Diamond | 83.5% | — | source ↗ | |
| 10 | Step-3.5-Flash Base | — | WinoGrande | 79.1% | — | source ↗ | |
| 11 | Step-3.5-Flash | — | SWE-Bench Verified | 74.4% | — | source ↗ | |
| 12 | Step-3.5-Flash Base | — | MATH | 66.8% | — | source ↗ | |
| 13 | Step-3.5-Flash | — | BrowseComp | 51.6% | — | source ↗ | |
| 14 | Step-3.5-Flash Base | — | GPQA Diamond | 41.7% | — | source ↗ | |
| 15 | Step-3.5-Flash Base | — | SimpleQA | 31.6% | — | source ↗ | |
| 16 | Step-3.5-Flash PaCoRe | — | HLE | 27.9% | — | source ↗ | |
| 17 | Step-3.5-Flash | — | HLE | 23.1% | — | source ↗ |
CodeSOTA extraction
Benchmark evidence
Link this paper to benchmark rows, datasets, model cards, and reproduced results as evidence is extracted.