English < Article List - DigitalToday

Search results for Opus 4.8

AI & Enterprise

U.S. tech firms increasingly use Chinese open-source AI models

Demand is rising for Chinese open-source AI models that are relatively cheaper while offering performance close to OpenAI and Anthropic, with interest spreading to major U.S. tech firms. Coinbase is testing Chinese open-weight models as defaults in its LLM gateway. Databricks’ Yu Chen Jin called GLM-5.2 an “open-source Claude moment.” Axios reported Microsoft is reviewing DeepSeek V4 fine-tuned and other open-source models as alternatives.

AI & Enterprise

Anthropic unveils Claude Sonnet 5, lowers price for Opus 4.8-level performance

Anthropic on Monday unveiled its mid-range model Claude Sonnet 5. The company said Sonnet 5 offers performance close to Opus 4.8 while being cheaper to use, and it is available as the default model for Claude Free and Pro plans and across all subscription tiers. Pricing through Aug. 31 is $2 per 1 million input tokens and $10 per 1 million output tokens.

AI & Enterprise

China AI matches Anthropic\'s Mythos in some security tasks, putting pressure on U.S. policy

Chinese AI systems are showing performance comparable to Anthropic’s top model Mythos in some cybersecurity scenarios, the Wall Street Journal reported. Researchers said Z.ai’s GLM-5.2, unveiled in June, matches U.S. models in bug detection and has ranked among the 10 most used models on OpenRouter data, at times beating Anthropic’s Claude Opus 4.8. The model is open-weight. China’s advances coincide with U.S. restrictions on model access.