Open-source LLM and dataset for sports forecasting (Pro Golf)
Hey HN, I fine-tuned a small open-source model on golf forecasting and it beats GPT-5 at predicting golf outcomes. The same approach can be used to build a specialized model in any domain, you just need to update a few search queries. We fine-tuned gpt-oss-120b with LoRA on 3,178 golf forecasting questions, using GRPO with Brier score as the reward. Our model outperformed GPT-5 on Brier Skill (17% vs 12.8%) and ECE (6% vs 10.6%) on 855 held-out questions. How to try it: the model and dataset are open-source, with code, on Hugging Face. How to build your own specialized model: Update the…
What it does
In the maker’s words, at launch
Hey HN, I fine-tuned a small open-source model on golf forecasting and it beats GPT-5 at predicting golf outcomes. The same approach can be used to build a specialized model in any domain, you just need to update a few search queries. We fine-tuned gpt-oss-120b with LoRA on 3,178 golf forecasting questions, using GRPO with Brier score as the reward. Our model outperformed GPT-5 on Brier Skill (17% vs 12.8%) and ECE (6% vs 10.6%) on 855 held-out questions. How to try it: the model and dataset are open-source, with code, on Hugging Face. How to build your own specialized model: Update the search queries and instructions in the Lightning Rod SDK to generate a new forecasting dataset, then run the same GRPO + LoRA recipe. SDK link: https://github.com/lightning-rod-labs/lightningrod-python-sd... Dataset: https://huggingface.co/datasets/LightningRodLabs/GolfForecas... Model: https://huggingface.co/LightningRodLabs/Golf-Forecaster Questions, feedback on the SDK, suggestions for new domains to try this on - all are welcome.
Does the same job
all alternatives →- GGGetting GLM 5.2 running on my slow computerJul 2026 · github.com · ▲937
A few days ago I found myself trying out GLM 5.2 and was really positively impressed. The capabilities and security I was getting from this LLM are similar to those I've gotten from models like Claude or GPT, and this really surprised me. But then I thought, "I wonder how it would work on a normal computer like mine," and above all, "I wonder if it would work without going into OOM on a computer like mine." So I started working with the help of agents to test this possibility. I started converting the model to int4, understanding MTP usage, and if possible implementing DSA for long context.…
- FTFine-tune an 8B model on a 4 GB laptop GPUAug 2026 · github.com · ▲139
- DDDistilling DeepSeek into GPT-OSS doesn't transfer censorship. Try itJul 2026 · ctgt.ai · ▲170
We recently used DeepSeek V4 Flash as a teacher for finance tasks with GPT-OSS-120B. Distillation works well on this problem. At a constrained 8k token budget, our self-distilled 120B scores 83.61% on FinanceReasoning, above Kimi K3 (81.93%) and Inkling (65.13%). We released the 20B open weights. With V4 as the teacher though, we realized it would be timely to measure if the censorship characteristic of it transferred to the distilled version of the base model. tl;dr it didn't, the teacher answered politically sensitive questions 7 SDs differently than expected, but the distilled model's…
- MLMinimal LLM Post-Training Experiments on an 8GB GPU (SFT, DPO, GRPO)Aug 2026 · github.com · ▲21
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
