OpenRouter

Evidence (6)

• 7Simon Willison·8/12/2026model_release
  DeepSeek V4 Pro 0813 released

  DeepSeek V4 Pro 0813 is a new model release listed on OpenRouter and discussed heavily on Hacker News. The post appears to center on availability and access to the model rather than a broader product announcement or research paper.

• 4Simon Willison·7/22/2026research
  Benchmarking Whether AI Labs Are "Pelicanmaxxing"

  An analysis tested whether major AI labs may be optimizing models for Simon Willison’s informal “pelican riding a bicycle” benchmark. The author generated 1,008 SVGs across seven frontier models, scored them with LLM judges, and found no clear evidence that pelican-on-bicycle outputs were better than other animal-vehicle combinations. The article provides the full experiment and code on GitHub.

• 7Hacker News·7/17/2026research
  Open source AI gains ground, but frontier gaps remain

  A Hacker News-linked CTO letter argues that open-weight AI has become competitive in many production workloads, with lower inference costs and broad adoption in tokens routed through open models. The report says closed models still lead on reasoning and multimodality, while open models are strong in coding, instruction-following, and general use cases.

• 8Simon Willison·7/6/2026model_release
  Tencent releases Hy3 open model

  Tencent has released Hy3, an Apache 2.0-licensed Mixture-of-Experts model with 295B total parameters and 21B active parameters. The company says it was improved with feedback from more than 50 products and now offers strong performance on utility and productivity tasks, with free access on OpenRouter until July 21.

• 6The New Stack·6/24/2026product_launch
  Sakana AI Launches Fugu, a Multi-Agent Orchestration System

  Sakana AI released Fugu, a multi-agent orchestration system that splits tasks into subtasks and routes them across swappable expert models through a single OpenAI-compatible API. The company says it can match frontier-model performance on several benchmarks, but early reactions question its pricing, burn rate, and whether it really advances AI sovereignty beyond an advanced routing layer.

• 8Simon Willison·6/17/2026model_release
  GLM-5.2 open weights model lands with 1M context

  Z.ai has released GLM-5.2 as open weights under an MIT license after an earlier limited rollout to coding subscribers. The text-only Mixture-of-Experts model has a 1 million token context window and is being reported as the leading open weights model on independent benchmarks, though it is relatively token-hungry.