Polish dev defends open model: built on Gemma with more than a simple LoRA fine-tune
ColinWPLewis · x · 2026-10-21
Responding to skeptics, a Polish developer explains why he open-sourced his model — commercially he's paid for other models, and this release showcases local AI capability. He confirms the flagship model is built on Gemma but insists it's not just a LoRA or plain fine-tune, inviting critics to train their own and benchmark it on JevBench; full details are in his latest video.
More from Models
- Stronger Model, Weaker Product: Dev Defaults to Fable 6.1 Sol Ultra Over 5.1 — curious_vii · 2026-10-21
- User claims switching from Claude Opus 5.5 to GPT-6.1 dramatically slowed his machine — koltregaskes · 2026-10-21
- Mercury 3 tested: the only competitive diffusion LLM, matches GPT-5.6 Luna on coding — volokuleshov · 2026-10-21
- Gemini APIs Add Blazing-Fast Live Speech Translation in Google AI Studio — DynamicWebPaige · 2026-10-21
- GLM-5.3-Flash NVFP4 benchmarked: 2x H200 matches 4x RTX Pro 6000 on throughput — TheZachMueller · 2026-10-21
- New ClinicalBench results raise doubts about both the benchmark and MiniMax's release — teortaxesTex · 2026-10-21