synoema-iot-lite-1.5b-v2

Language 0.1.0-beta.1

Code generator, not a chatbot. Give it an IoT automation task — it outputs Synoema code. Example: Turn on fan when temperature exceeds 30

Trained on Synoema — a formally verified functional language (GBNF + Hindley-Milner + contracts, prompt to native/WASM/IoT with no human review).

Evaluation

Metric Value
compile_pass 97.1%
run_pass 57.7%
Eval set 104 examples
Method greedy (do_sample=False)
Language version 0.1.0-beta.1

Quickstart

wget https://huggingface.co/delimitter/synoema-iot-lite-1.5b-v2/resolve/main/synoema-iot-lite-1.5b-v2-q4km.gguf
wget https://huggingface.co/delimitter/synoema-iot-lite-1.5b-v2/resolve/main/Makefile
make pull
make run

make pull sets the system prompt automatically — required for correct behavior.

Target Action
make pull Create Ollama model + system prompt
make run Interactive chat
make clean Remove from Ollama

Example prompts

Turn on fan when temperature exceeds 30
humidity > 80 -> turn on pump on GPIO 22
Turn off lights when no motion for 10 minutes
Open valve when pressure exceeds 150

Model details

Field Value
Base model Qwen/Qwen2.5-Coder-1.5B-Instruct
Fine-tuning QLoRA SFT (LoRA r=16)
Size 941 MB Q4_K_M
Language version 0.1.0-beta.1

Links

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for delimitter/synoema-iot-lite-1.5b-v2

Finetuned
(160)
this model