Model card
MambaVision-L2.
NVIDIAopen-source241M paramsHybrid Mamba-Transformer
Hybrid Mamba-Transformer vision backbone. Large-2 variant with 241M params. Trained from scratch on ImageNet-1K at 224px. CVPR 2025. Paper: arxiv:2407.08083.
§ 02 · Benchmarks
Every benchmark MambaVision-L2 has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | ImageNet-1K | Computer Vision · Image Classification | top-1-accuracy | 85.3% | #10 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area
Where MambaVision-L2 actually performs.
§ 05 · Related models
Other NVIDIA models scored on Codesota.
§ 06 · Sources & freshness
Where these numbers come from.
cvpr-2025
1
result
0 of 1 rows marked verified.