Models / Qwen2.5 / Qwen2.5 3B
SamTheDev/ qwen2.5-3B-ogaText generation · served with ONNX Runtime GenAI · ↓ 0 downloads a month · View on Hugging Face
Weights 2.4 GB
GPU memory needed about 4.5 GB
Memory · disk 8 GB · 36 GB
Licence mit Versions Smaller versions (8-bit, 4-bit) need less GPU memory and answer a little less precisely.
Versions of Qwen2.5 3B The same model packaged differently: a quantized version needs less GPU memory and runs on a cheaper machine.
Version Runtime Weights GPU memory Cheapest machine now Original Qwen/Qwen2.5-3B vLLM 6.2 GB 9.4 GB $0.10/h RTX 3060 Instruct Qwen/Qwen2.5-3B-Instruct vLLM 6.2 GB 9.4 GB $0.10/h RTX 3060 Instruct AWQ Qwen/Qwen2.5-3B-Instruct-AWQ vLLM 2.7 GB 5.5 GB $0.10/h RTX 3060 Qwen2.5 Coder 3B Instruct Qwen/Qwen2.5-Coder-3B-Instruct vLLM 6.2 GB 9.4 GB $0.10/h RTX 3060 Qwen2.5 Coder 3B Qwen/Qwen2.5-Coder-3B vLLM 6.2 GB 9.4 GB $0.10/h RTX 3060 Qwen2.5 Coder 3B Instruct AWQ Qwen/Qwen2.5-Coder-3B-Instruct-AWQ vLLM 2.7 GB 5.5 GB $0.10/h RTX 3060
Hardware and deploy Cheapest GTX 1660 SUPER 6 GB GPU memory from $0.07/h Best value RTX 5090 32 GB GPU memory 116x the speed of the cheapest from $0.62/h Fastest Nvidia B200 179 GB GPU memory 400x the speed of the cheapest from $6.89/h
Choose another machine (54 fit this model)
Deploying ONNX models from the catalog is coming soon. Today, download the files from Hugging Face and upload them in Inference.
Prices are per hour for the whole machine and its disk, right now. Speed is Vast's DLPerf score of the machine, relative to the cheapest. Best value: the most speed per dollar. Only machines with a Vast reliability above 95% that download at 400 Mbit/s or more are offered: the machine fetches the model itself.