MANAS-1
NeuroDX announces India's first Brain Language Model, MANAS-1. Designed as a Foundation Model for EEGs, MANAS-1 demonstrates SOTA performance across various tasks in BCI and clinical applications. We have released two variants, MANAS-1-small (40M parameters), and MANAS-1-medium (400M parameters).
Benchmarking Results:
EEG-FM-Bench
These metrics were computed using the EEG-FM-Bench codebase, using linear probing.
| Model | bcic_2a | adftd | workload | motor_mv_img | siena_scalp | tuar | tusl |
|---|---|---|---|---|---|---|---|
| CSBrain | 28.5 +/- 1.81 | 47.2 +/- 2.36 | 53.0 +/- 2.09 | 26.4 +/- 0.78 | 50.0 +/- 0.00 | 27.2 +/- 2.12 | 54.6 +/- 6.86 |
| CBraMod | 36.4 +/- 0.43 | 37.5 +/- 2.40 | 48.5 +/- 0.41 | 52.4 +/- 0.29 | 67.8 +/- 1.63 | 27.4 +/- 0.32 | 44.5 +/- 1.77 |
| BIOT | 24.8 +/- 1.69 | 42.7 +/- 4.11 | 51.9 +/- 2.56 | 27.0 +/- 0.62 | 64.0 +/- 1.66 | 27.1 +/- 0.51 | 55.9 +/- 6.78 |
| REVE | 32.5 +/- 0.24 | 34.2 +/- 5.15 | 65.5 +/- 1.75 | 45.9 +/- 1.07 | 61.4 +/- 3.33 | 25.3 +/- 1.89 | 55.6 +/- 3.95 |
| MANAS-1-small | 49.5 +/- 1.00 | 52.9 +/- 2.94 | 60.5 +/- 2.26 | 60.4 +/- 0.21 | 68.7 +/- 1.77 | 35.7 +/- 3.90 | 62.1 +/- 5.43 |
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support