Models / ultravox-v0_5-llama-3_2

ultravox-v0_5-llama-3_2 by fixie-ai

GPU memory needed (smallest version of each): ultravox-v0_5-llama-3_2 1B about 1.4 GB. The cheapest runs on Quadro P4000 from $0.07/h right now.

1 model (1B) · ↓ 78.8K downloads a month · from $0.07/h on Quadro P4000 right now. Each model's versions do the same job; quantized ones (AWQ, FP8, 4-bit) need less GPU memory and run on cheaper machines.

ultravox-v0_5-llama-3_2 1B

Version	Runtime	Weights	GPU memory	Cheapest machine now
Originalfixie-ai/ultravox-v0_5-llama-3_2-1b	vLLM	1.4 GB	6.1 GB	$0.09/h RTX 3060
ONNX (onnx-community)onnx-community/ultravox-v0_5-llama-3_2-1b-ONNX	ONNX Runtime	260 MB	1.4 GB	$0.07/h Quadro P4000

Model details come from Hugging Face and are refreshed daily; prices follow the live offers. GPU memory is GridPup's estimate: the weights, room for an 8,192-token conversation (chat models) and the runtime's own memory. Large models are split across the GPUs of one machine ("4x RTX 4090").

Loading…