JetBrains has launched Mellum2, a 120-billion parameter Mixture-of-Experts model, detailed on Hugging Face. Separately, Hugging Face has published a collection of past evaluation results for various models on its model page. Additionally, IBM Research has introduced ScarfBench, a benchmark for AI agents focused on migrating enterprise Java frameworks. AI
IMPACT New model releases and benchmarks from major entities like JetBrains and IBM Research indicate continued advancements in AI capabilities and evaluation methodologies.
RANK_REASON Cluster includes a new frontier model release (Mellum2) from JetBrains.
Read on Mastodon — sigmoid.social →
- AI agents
- eee-community-evals
- Enterprise Java Framework Migration
- Hugging Face
- IBM Research
- JetBrains
- Mellum2
- ScarfBench
AI-generated summary · Google Gemini · from 3 sources. How we write summaries →