Soofi S 30B-A3B releases a full pretraining report and claims open-model leads in English and German
abursuc · x · 2026-07-21
Soofi S 30B-A3B publishes a full pretraining report, claiming top open-model scores in English and German
The project released the full pretraining tech report and project page for Soofi S 30B-A3B, a Mixture-of-Experts hybrid Mamba model trained on 27 trillion tokens with German deliberately upweighted.
What the report says
- The team says it is the strongest fully open model in its evaluations on both the English and German aggregates, ahead of Olmo 3 32B and Apertus 70B.
- The release emphasizes radical transparency: complete per-source data accounting, all hyperparameters, training and evaluation code, and checkpoints under permissive licenses.
- The model was trained end-to-end on Deutsche Telekom infrastructure.
The thread’s criticism
The quoted reply pushes back hard on the framing, arguing that loudly claiming “sovereignty” and “champion” status creates confusion and misleads the public, especially around German/EU research positioning.
More from Models
- Testing Qwen Image Model: One Prompt Generates Maps Pleasing Both India and Pakistan — NirantK · 2026-07-22
- Gemini 3.5 Flash Lite Tested: Not Frontier-Optimal, but Hits 350 tok/s — brandon_galang · 2026-07-22
- Benchmark scorecard pits GPT-5.6 Sol, Claude Fable 5, and Gemini 3.6 Flash — iruletheworldmo · 2026-07-22
- Google says Gemini 3.5 Pro is testing with partners and will go wide when ready — inductionheads · 2026-07-22
- ChatGPT merges Codex and adds GPT-5.6 tiers for knowledge work — jxnlco · 2026-07-22
- Dev Complains About Google's Fast Model Deprecation: Frequent Swaps Increase Maintainability Debt — Justin_Halford_ · 2026-07-22