MiniCPM5-1B ONNX Web

ONNX export of openbmb/MiniCPM5-1B for browser/runtime experiments.

Export

  • Source: openbmb/MiniCPM5-1B
  • Architecture: LlamaForCausalLM
  • ONNX opset: 18
  • Inputs: input_ids, attention_mask
  • Output: logits
  • KV cache: disabled for the first broad-compatibility export
  • External data: yes, weights are stored in model.onnx.data

This artifact is valid ONNX and passes onnx.checker.check_model. It is intentionally a first browser-deployment baseline; production browser serving should add a smaller quantized/sharded artifact after runtime compatibility testing.

Downloads last month
17
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Reza2kn/MiniCPM5-1B-ONNX-Web

Quantized
(19)
this model