Benchmark
AI benchmark news — how leading models score on reasoning, coding, and other tests, and what the results actually mean.
6 articles
BenchmarkXiaomi Open-Sources Xiaomi-Lens-0, a 4B Model That Points to Image Regions Pixel by Pixel
Matthieu Morel · 21 Oct 2026 · 14 min read
BenchmarkA 27B Quantized LLM Is Said To Match Frontier AI Models In Just One Task From A Coding Benchmark, Making It A More Belie
Matthieu Morel · 2 Oct 2026 · 12 min read
Benchmarkmemoket/memoket-kite: An open-source, vector-free long-term memory engine for AI agents, achieving SOTA on LoCoMo and Lo
Matthieu Morel · 26 Aug 2026 · 13 min read
BenchmarkBenchmarking Coding Agents on Databricks’ Multi-Million Line Codebase
Matthieu Morel · 10 Jul 2026 · 13 min read
BenchmarkHow has your AI workflow changed over the past year?
Matthieu Morel · 1 Jul 2026 · 11 min read
BenchmarkOpen-source single-GPU reproductions of Cartridges and STILL for neural KV-cache compaction [P]
Matthieu Morel · 15 Jun 2026 · 12 min read