News
Latest news and updates on local artificial intelligence
2 news items in Benchmarks
Co-Design Bench: 273 tasks from real user requests to evaluate multimodal coding agents
November 5, 2026A new reference-free Benchmark: 273 tasks and 2,331 expert-authored acceptance criteria, curated from 37,463 deduplicated queries across 26,314 real user coding sessions; two tracks (100 multimodal editing tasks, 173 repository editing tasks) and a judge-then-refute protocol that evaluates the running application with browser evidence.
BenchLM publishes its September ranking of the best local LLMs: VRAM defines the tiers
September 27, 2026The September 2026 ranking orders Open-weight models by VRAM tier: GPT-OSS 20B and ZAYA1-8B lead the 8-16 GB tier, Qwen3.6-27B (46.49) and Gemma 4 31B (44.84) dominate the 16-24 GB tier on a single RTX 4090 at Q4, and trillion-Parameter MoEs need 8x H100. A useful reference when choosing a model size for your hardware.