GPU Vault model library

Find a model. Launch it on the right GPU.

Browse currently supported Hugging Face and official ComfyUI models, compare the hardware profile, and review a live GPU offer before spending credits.

Sort

Launchable models

16 launchable models matching “RedHatAI” on this page. Exact, runtime-qualified revisions only.

text generationRedHatAI

Llama-3.2-3B-Instruct-FP8-dynamic

Llama-3.2-3B-Instruct-FP8-dynamic is a validated text generation model.

1× GPU · 24 GB VRAM3.6B params 361.5K

text generationRedHatAI

Llama-3.2-1B-Instruct-FP8-dynamic

Llama-3.2-1B-Instruct-FP8-dynamic is a validated text generation model.

1× GPU · 16 GB VRAM1.5B params 296.5K

text generationRedHatAI

Qwen3-32B-NVFP4

Qwen3-32B-NVFP4 is a validated text generation model.

1× GPU · 40 GB VRAM19.1B params 270.9K

text generationRedHatAI

Llama-3.2-3B-Instruct-FP8

Llama-3.2-3B-Instruct-FP8 is a validated text generation model.

1× GPU · 24 GB VRAM3.6B params 184.2K

text generationRedHatAI

Qwen2.5-1.5B-quantized.w8a8

Qwen2.5-1.5B-quantized.w8a8 is a validated text generation model.

1× GPU · 16 GB VRAM1.8B params 162.4K

text generationRedHatAI

Meta-Llama-3.1-70B-Instruct-FP8

Meta-Llama-3.1-70B-Instruct-FP8 is a validated text generation model.

1× GPU · 96 GB VRAM70.6B params 139.2K

text generationRedHatAI

Meta-Llama-3.1-8B-Instruct-FP8

Meta-Llama-3.1-8B-Instruct-FP8 is a validated text generation model.

1× GPU · 24 GB VRAM8B params 135.6K

text generationRedHatAI

Llama-3.3-70B-Instruct-FP8-dynamic

This model was obtained by quantizing activation and weights of Llama-3.3-70B-Instruct to FP8 data type.

1× GPU · 96 GB VRAM70.6B params 119.4K

text generationRedHatAI

phi-4-FP8-dynamic

This model was obtained by quantizing activation and weights of phi-4 to FP8 data type.

1× GPU · 40 GB VRAM14.7B params 100.2K

text generationRedHatAI

Llama-3.2-1B-Instruct-FP8

Llama-3.2-1B-Instruct-FP8 is a validated text generation model.

1× GPU · 16 GB VRAM1.5B params 80.2K

text generationRedHatAI

Meta-Llama-3.1-8B-Instruct-quantized.w4a16

This model was obtained by quantizing the weights of Meta-Llama-3.1-8B-Instruct to INT4 data type.

1× GPU · 48 GB VRAM8B params 50.2K

text generationRedHatAI

DeepSeek-R1-Distill-Qwen-32B-quantized.w4a16

DeepSeek-R1-Distill-Qwen-32B-quantized.w4a16 is a validated text generation model.

2× GPU · 80 GB VRAM32.8B params 16.8K

text generationRedHatAI

Llama-3.3-70B-Instruct-NVFP4

Llama-3.3-70B-Instruct-NVFP4 is a validated text generation model.

1× GPU · 80 GB VRAM40.6B params 14.4K

text generationRedHatAI

Meta-Llama-3.1-8B-Instruct-FP8-dynamic

This model was obtained by quantizing the weights and activations of Meta-Llama-3.1-8B-Instruct to FP8 data type.

1× GPU · 24 GB VRAM8B params 13.7K

text generationRedHatAI

Qwen3-0.6B-quantized.w4a16

Qwen3-0.6B-quantized.w4a16 is a validated text generation model.

1× GPU · 16 GB VRAM765.4M params 12.6K

text generationRedHatAI

Qwen3-8B-quantized.w4a16

Qwen3-8B-quantized.w4a16 is a validated text generation model.

1× GPU · 48 GB VRAM8.3B params 12.5K