The only fine tune I've attempted is a Qwen3.5:0.8b trained on RL data generated using Qwen3.5:122b playing 50-thousand hands of Hearts (the card game) because

The only fine tune I've attempted is a Qwen3.5:0.8b trained on RL data generated using Qwen3.5:122b playing 50-thousand hands of Hearts (the card game) because I wanted a tiny model that was, idealy, an expert Hearts player so I could hopefully get a super-fast local model that I could play hearts against without any per-turn lag. I wanted more than 50k hands but even just 50k took almost 3 days of RL running on a RunPod RTX 4090. Used the RL data to build a LoRa adapter for Qwen3.5:0.8b -- ultimately a failure but mostly because the data we generated wasn't of sufficient quality. Turns out hearts is a very complex game as played by sentient humans lol.