337 summarised stories about Open-Weight Models, each linking back to the original source. Browse all topics →
Amazon Web Services 1 month ago 12
AWS made the open-weight Moonshot AI Kimi K3 model available on Amazon Bedrock for coding and knowledge work. The model is first open to reach 2.8 trillion parameters and supports a 1-million-token context window. It brings features like explicit prompt caching and updated Bedrock inference APIs and data-protection settings to new deployments without changing customers’ security posture.
The New Stack 1 month ago 38 ● 2 sources
Open-weight models took a majority of tokens routed through Vercel’s AI Gateway, reaching 56% of monthly token volume for the first time. Anthropic still accounted for 64% of the estimated dollars spent via the gateway in August, even though open-weight models processed 56% of the tokens. As open-weight usage grows and average per-token pricing declines (down 23.2% in August), spend remains concentrated with Anthropic while teams shift within and across model labs based on cost and consistency.
The daily briefing
TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.
TLDRocket
22 October 2026
The biggest thread today is the push to make AI agents reusable in organizations, not just clever in demos. Sakana AI’s “Sakana Aquarium” leans on ontology as the shared backbone for decision-making and execution, aiming to let teams spin up new agents by reusing prepared data and a maintained knowledge model instead of starting from scratch each workflow. That direction keeps showing up in adjacent efforts to standardize how agents package skills and access tools: the Agentic AI Foundation added the Agent Skills open standard to sit under the Linux Foundation, with 70-plus platforms integrating since December 2025 and hundreds of millions of installations behind them. Even Anthropic’s new open call for life-sciences “guest investigators” using Claude fits the same pattern—8 to 12 weeks of collaboration designed to produce publishable, open artifacts while leaving researchers in control of their own data.
Read the full briefing →