Oya: wallet agent models
Collection
Tiny Aya fine-tuned into a tool-calling wallet agent for English, Nigerian Pidgin and Yoruba (Lexsi Labs hackathon, Track 3). โข 5 items โข Updated
How to use timidan/oya-tiny-aya-gguf with llama.cpp:
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf timidan/oya-tiny-aya-gguf:Q4_K_M # Run inference directly in the terminal: llama cli -hf timidan/oya-tiny-aya-gguf:Q4_K_M
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf timidan/oya-tiny-aya-gguf:Q4_K_M # Run inference directly in the terminal: llama cli -hf timidan/oya-tiny-aya-gguf:Q4_K_M
git clone https://github.com/ggml-org/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-app # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama serve -hf timidan/oya-tiny-aya-gguf:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama cli -hf timidan/oya-tiny-aya-gguf:Q4_K_M
docker model run hf.co/timidan/oya-tiny-aya-gguf:Q4_K_M
How to use timidan/oya-tiny-aya-gguf with Ollama:
ollama run hf.co/timidan/oya-tiny-aya-gguf:Q4_K_M
How to use timidan/oya-tiny-aya-gguf with Docker Model Runner:
docker model run hf.co/timidan/oya-tiny-aya-gguf:Q4_K_M
How to use timidan/oya-tiny-aya-gguf with Lemonade:
# Download Lemonade from https://lemonade-server.ai/ lemonade pull timidan/oya-tiny-aya-gguf:Q4_K_M
lemonade run user.oya-tiny-aya-gguf-Q4_K_M
lemonade list
CohereLabs/tiny-aya-global fine-tuned with LoRA (sft-v2) for tool calling in English, Nigerian Pidgin and Yoruba,
merged and quantized with llama.cpp for CPU inference.
Prompt format: compact style from https://github.com/Timidan/lexsi-hack (naija_agent/prompting.py).
Demo: https://huggingface.co/spaces/timidan/oya
Built with Tiny Aya by Cohere Labs (CC-BY-NC 4.0; Cohere Labs Acceptable Use Policy applies).
4-bit