All
Text
Vision
Audio
Multimodal
| Model | Category | Specs | Download |
|---|---|---|---|
| PulseMini-2 Streaming text inference tuned for low-latency chat. | Text · Real-time Transformer · 1.2B params |
Latency: 28ms @ A10G Context: 32k tokens |
Download |
| AtlasFrame Understands interfaces, charts, and dense text overlays. | Multimodal · Vision+Text Hybrid encoder · 6.5B params |
Latency: 120ms @ A100 Context: 8 images + 8k tokens |
Download |
| EchoWeave Generates lifelike narration with adjustable pacing. | Audio · Generation Diffusion · 900M params |
Latency: 2.1s / min audio Voices: 60 multilingual presets |
Download |
| VectorPulse Sentence embedding model optimized for semantic search. | Text · Embeddings Transformer · 800M params |
Dimensionality: 2048 Throughput: 2k qps / GPU |
Download |
| ScenePilot Vision-language planner for robotics and spatial navigation. | Multimodal · Control Transformer + policy head |
Latency: 180ms @ V100 Inputs: RGB-D + natural language |
Download |
Drop your packaged model files inside
downloads/ and update the file names above. Links include the download attribute so browsers save the archive immediately.