Compare LLM Models
Select 2-5 models to compare side by side.
Select models to compare
0 / 5 selected
Top downloads
Select 2-5 models
| Attribute |
inclusionAI
|
|---|---|
| Company | inclusionAI |
| Release Date | 2026-09-04 |
| Parameters (B) | 124 |
| Architecture | Hybrid MoE Vision-Language Transformer (ViT encoder + KDA/Gated MLA backbone) |
| Context Window | 262144 |
| Input Modalities | text, image, video |
| Open Source | Yes |
| License | mit |
| Score | 25 |
| VRAM (GB) | 80 |
| Compute | 4x NVIDIA H100 80GB (BF16, 256K context); scale to 8x H100 on 80GB cards |
| Benchmark: GPQA-Diamond | 86.2 |
| Benchmark: HLE | 22 |
| Benchmark: MMMU-Pro | 79 |
| Benchmark: Terminal-Bench-2.1 | 64.4 |
| Pricing (Input) | 0.02 |
| Pricing (Output) | 0.06 |