TitleOS/Nemotron-3-Ultra-High-Effort-OpenCode-Distilled-2k
Just under 2000 CoT synthetic prompt chains distilled from Nemotron 3 Ultra w/ High effort gathered using OpenCode entirely for free. Using a propertary harness developed by me but currently private, partially synthetic or fully sythentic datasets can be created from any of the free models offered by OpenCode, with respect to rate-limiting. In this case, a fully sythentic dataset of 1800ish rows was generated by "nemotron-3-ultra-free" using the maximium high effort reasoning. This was done by… See the full description on the dataset page: https://huggingface.co/datasets/TitleOS/Nemotron-3-Ultra-High-Effort-OpenCode-Distilled-2k.
Just under 2000 CoT synthetic prompt chains distilled from Nemotron 3 Ultra w/ High effort gathered using OpenCode entirely for free. Using a propertary harness developed by me but currently private, partially synthetic or fully sythentic datasets can be created from any of the free models offered by OpenCode, with respect to rate-limiting. In this case, a fully sythentic dataset of 1800ish rows was generated by "nemotron-3-ultra-free" using the maximium high effort reasoning. This was done by using the full sythentic "generate" mode of my OpenReasoningHarness, which prompts the model for the user's prompt, then completes it in fresh context, with CoT preserved. I will attempt to honor requests for other models to be distilled, the following are available for free on OpenCode as of 6/25/26:
Free models (registry key -> label):
deepseek-v4-flash-free DeepSeek V4 Flash Free effort: low/medium/high/max *default
opencode-zen OpenCode Zen effort: low/medium/high
big-pickle Big Pickle effort: low/medium/high
mimo-v2.5-free MiMo V2.5 Free effort: low/medium/high
nemotron-3-ultra-free Nemotron 3 Ultra Free effort: low/medium/high/max
north-mini-code-free North Mini Code Free effort: low/medium/high