How much GPU memory does gigaam-v3-onnx need?

About 2.3 GB: 0.89 GB of weights and the runtime's own memory.

What is the cheapest GPU to run gigaam-v3-onnx?

Right now one Quadro P4000 at $0.07 an hour; RTX 5090 at $0.53 an hour is the best value for its speed. Prices are live offers from GPU clouds, checked every 15 minutes.

Can gigaam-v3-onnx run on an RTX 4090?

Yes: it needs about 2.3 GB and an RTX 4090 has 24 GB.

Models / GigaAM

istupakov/gigaam-v3-onnx

Speech to text · served with ONNX Runtime · ↓ 8K downloads a month · View on Hugging Face

istupakov/gigaam-v3-onnx needs about 2.3 GB of GPU memory. It runs on one Quadro P4000 from $0.07/h right now; best value: RTX 5090 at $0.53/h.

These ONNX files are an export for running in a browser (an audio encoder and a text decoder, each a version here): served as tensors, they do not make a transcript. For speech to text, pick the original model (for example openai/whisper-base), served with vLLM.

Weights: 890 MB
GPU memory needed: about 2.3 GB
Memory · disk: 8 GB · 30 GB
Licence: mit

Versions

Smaller versions (8-bit, 4-bit) need less GPU memory and answer a little less precisely.

8-bitv3_ctc.int8.onnx220 MB · 1.3 GB GPU full precisionv3_ctc.onnx890 MB · 2.3 GB GPU 8-bitv3_e2e_ctc.int8.onnx220 MB · 1.3 GB GPU full precisionv3_e2e_ctc.onnx890 MB · 2.3 GB GPU 8-bitv3_e2e_rnnt_decoder.int8.onnx1 MB · 1 GB GPU full precisionv3_e2e_rnnt_decoder.onnx1 MB · 1 GB GPU 8-bitv3_e2e_rnnt_encoder.int8.onnx220 MB · 1.3 GB GPU full precisionv3_e2e_rnnt_encoder.onnx890 MB · 2.3 GB GPU 8-bitv3_e2e_rnnt_joint.int8.onnx1 MB · 1 GB GPU

Hardware and deploy

Deploying ONNX models from the catalog is coming soon. Today, download the files from Hugging Face and upload them in Inference.

Questions

How much GPU memory does gigaam-v3-onnx need?: About 2.3 GB: 0.89 GB of weights and the runtime's own memory.
What is the cheapest GPU to run gigaam-v3-onnx?: Right now one Quadro P4000 at $0.07 an hour; RTX 5090 at $0.53 an hour is the best value for its speed. Prices are live offers from GPU clouds, checked every 15 minutes.
Can gigaam-v3-onnx run on an RTX 4090?: Yes: it needs about 2.3 GB and an RTX 4090 has 24 GB.

Loading…