… These developments lead to strong performance in multimodal coding, visual tool use , and framework-based agentic tasks, while preserving competitive text-only coding capability. …
… The 7.2B model achieves ~1.6× decoding throughput speedup over baselines while staying competitive on accuracy — a promising direction for faster, scalable generation. …
… The following papers were recommended by the Semantic Scholar API Teaching Thinking Models to Reason with Tools: A Full-Pipeline Recipe for Tool-Integrated Reasoning 2026 Think Anywhere in Code Generation 2026 HintMR: Eliciting Stronger Mathematical Reasoning in Small Language Models 2026 Efficient… …
… Experiments across understanding, generation, long-text rendering, and editing benchmarks show that JoyAI-Image achieves state-of-the-art or highly competitive performance. …