Back to home
fine-tuning
3 articles tagged with this topic
QwenSolidity
Weekend Solidity Fine-Tune Beats Opus: Vertical Small Models' ROI Moment
A developer fine-tuned Qwen into a 27B Solidity model, beating Claude Opus on coding benchmarks. The signal: cheap small vertical models are catching
May 62 min read
Gemma-4Google-DeepMind
DeepMind’s New AI: A Gift To Humanity
Google DeepMind has released Gemma 4, a new family of open-weight models available under the Apache 2.0 license.
Apr 163 min read
Llamafine-tuning
Fine-Tuning on 4chan Data Boosts Llama 8B and 70B Benchmark Scores
A researcher fine-tuned Llama 8B and 70B on 4chan data and reports both models outperformed their base versions.
Apr 62 min read