SWE-bench-Live
PulseAugur coverage of SWE-bench-Live — every cluster mentioning SWE-bench-Live across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
Study finds AI patch performance differs from correctness rates
A new study published on arXiv explores the performance of AI-generated code patches, finding that correctness alone is insufficient to evaluate LLM capabilities in software development. Researchers repurposed existing …
-
New Tiel-Coder 35B model excels at coding and long conversations
A new open-source model, Tiel-Coder-35B-A3B, has been released, optimized for coding tasks and long conversations. It achieves strong performance on the SWE-bench-Live benchmark, fixing 12 out of 25 problems, which is c…
-
New Qwen Chat Template Boosts Coding and Knowledge Work Efficiency
A new chat template called Qwen-Sharp-Chat-Templates has been released, designed to optimize Qwen models for knowledge work and coding. This template aims to make models like Qwen3.8-27b more intelligent and efficient b…