| Feature | Name | Packaging | Model Format | Inference Backend | Hardware Support | ||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Language | Llama3.2-3B-Instruct | IGI Plugin | ONNX | DML | Intel, AMD, NVIDIA GPUs or CPUs | ||||||||||||||||||||
| Language | Llama3.2-3B-Instruct | NIM | TensorRT | TensorRT-LLM | Cloud, On-prem or Local GPUs | ||||||||||||||||||||
| Language | Qwen3 Family (.6B, 4B, 8B) | IGI Plugin | GGUF | GGML | Intel, AMD, NVIDIA GPUs or CPUs | ||||||||||||||||||||
| Language | Mistral Nemo Family (2B, 4B, 8B) | IGI Plugin | GGUF | GGML | Intel, AMD, NVIDIA GPUs or CPUs | ||||||||||||||||||||
| Language | Mistral Nemo Family (2B, 4B, 8B) | NIM | .nemo checkpoint | TensorRT-LLM | Cloud, On-prem or Local GPUs | ||||||||||||||||||||
| Language | Nemotron 3 Nano 4B | IGI Plugin | GGUF | GGML | Intel, AMD, NVIDIA GPUs or CPUs | ||||||||||||||||||||
| Language | Nemotron Nano 9B V2 | IGI Plugin | GGUF | GGML | Intel, AMD, NVIDIA GPUs or CPUs | ||||||||||||||||||||
| Vision | Nemovision 4B Instruct | IGI Plugin | GGUF | GGML | Intel, AMD, NVIDIA GPUs or CPUs | ||||||||||||||||||||
| Memory | E5 Large Unsupervised Embedded RAG | IGI Plugin | GGUF | GGML | Intel, AMD, NVIDIA GPUs or CPUs | ||||||||||||||||||||
| Speech | Riva ASR Family (140M, 600M) | IGI Plugin | ONNXTRT | DMLTRT + Triton | Intel, AMD, NVIDIA GPUs or CPUs | ||||||||||||||||||||
| Speech | Riva ASR NIM | NIM | RMIR | Triton + TRT | Cloud, On-prem or Local GPUs | ||||||||||||||||||||
| Speech | Resemble AI Chatterbox Turbo 350M TTS | IGI Plugin | GGUF | GGML | Intel, AMD, NVIDIA GPUs or CPUs | ||||||||||||||||||||
| Speech | Riva Magpie Flow TTS (FP16 and Q4) | IGI Plugin | ONNX | DML + TRT | Intel, AMD, NVIDIA GPUs or CPUs | ||||||||||||||||||||
| Speech | Whisper ASR | IGI Plugin | GGUF | GGML | Intel, AMD, NVIDIA GPUs or CPUs | ||||||||||||||||||||
| Speech | Whisper ASR | NIM | ONNX | Triton + TRT | Cloud, On-prem or Local GPUs |
1 Like