Math Solution App: on-device models

Model files downloaded by the Math Solution App (Android). Everything runs offline on the phone. models.json lists every pack, file, size and SHA-256 checksum.

Pack Files Size Runtime Licence
Question Reader unimernet-small-int8/ 209 MB ONNX Runtime Apache-2.0
AI Solver deepseek-r1-distill-qwen-1.5b-q4_k_m.gguf 1117 MB llama.cpp MIT

Sources and changes

  • UniMERNet-small by OpenDataLab (GitHub, Apache-2.0). Converted from PyTorch to ONNX (encoder + KV-cached decoder) and INT8 weight-quantised. Output was verified to match the original model.
  • DeepSeek-R1-Distill-Qwen-1.5B by DeepSeek (MIT), GGUF Q4_K_M quantisation, unmodified.

Full licence texts are in licenses/.

Downloads last month
-
GGUF
Model size
2B params
Architecture
qwen2
Hardware compatibility
Log In to add your hardware

4-bit

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