Topics

Topic

Claude Opus

5 written stories · 7 tracked

Tech3 outletsover 18h

Sources: Moonshot plans to launch Kimi K3, China's largest model to date with 2T-3T parameters, in the coming days; it's expected to outperform Claude Opus 4.8 (Financial Times)

Moonshot AI has released Kimi K3, a 2.8-trillion-parameter model that it describes as the largest open-source AI system to date. The model is positioned as a rival to leading U.S. offerings such as Claude Opus 4.8 and GPT 5.5, and the company plans to make the model weights publicly available by late July.

TechMeme·VentureBeat·The Next Web
Tech3 outlets

Anthropic’s Claude Opus 4.8 is its most honest AI model yet, and Mythos is coming in weeks

Anthropic has launched Claude Opus 4.8, featuring a new “dynamic workflow” tool and a fast mode that is three times cheaper than previous offerings. The model is described as the most honest version to date, with alignment approaching that of the upcoming Mythos system, which is expected to arrive in the coming weeks.

TechCrunch·VentureBeat·The Next Web
Tech3 outlets

Figma stock drops ~7% after Anthropic launched Claude Design, a dedicated app powered by its latest model Claude Opus 4.7 (Jon Keegan/Sherwood News)

Anthropic has launched Claude Design, an experimental AI product that generates visuals such as prototypes, slides, one-pagers and other quick designs using its Claude Opus 4.7 model. The tool is positioned as a dedicated app that directly competes with Figma, and following its announcement Figma’s stock fell roughly seven percent.

VentureBeat·TechCrunch·TechMeme
Tech3 outlets

Anthropic releases Claude Opus 4.7, narrowly retaking lead for most powerful generally available LLM

Anthropic has launched Claude Opus 4.7, describing it as a notable improvement over Opus 4.6 in advanced software-engineering tasks and adding a new “xhigh” effort level. The model uses an updated tokenizer that enhances text processing, though the same input may generate more tokens, and it delivers benchmark-leading coding and agentic performance, narrowly reclaiming the lead as the most powerful generally available large language model.

TechMeme·The Next Web·VentureBeat
Tech7 outletsover 30h

Anthropic's Claude Opus 4.6 brings 1M token context and 'agent teams' to take on OpenAI's Codex

Anthropic has released Claude Opus 4.6, a model with a 1 million-token context window that adds “agent teams” capable of breaking larger tasks into smaller jobs and integrates directly into PowerPoint via a side panel. The upgrade is marketed for enterprise use, excelling in coding, tool use, financial analysis and security testing, and is offered at the same pricing as its predecessor.

VentureBeat·TechCrunch·TechMeme·The Verge·Hacker News+2