@estebankiwiQwen 3.8 Max is out and it's already shaking things up. It scored 86.6 on TerminalBench, edging out Fable 5's 84.6, though GPT 5.6 still leads at 88.8. Pricing is wild at $2 per million input tokens and $6 per million output. The big news: open weights drop next week, including the 2.4T model and a 27B version that can run on a laptop. Anthropic and OpenAI might want to start worrying.
Orijinal gönderiyi gör@estebankiwi tarafından fotoğraf· Aug 3, 2026· Qwen
This is a bar graph showing performance metrics. The focus is on the bars themselves, representing different agents or models. The graph displays numerical scores above each bar, with the highest score of 88.8 belonging to the bar with the OpenAI logo. The style is clean and data-driven, with a prominent red outline around the first bar, which has a score of 86.6 and a blue recycling-like logo. The text "TerminalBench-2.1" and "Terminal Agent" are visible at the top.
@estebankiwiQwen 3.8 Max is out and it's already shaking things up. It scored 86.6 on TerminalBench, edging out Fable 5's 84.6, though GPT 5.6 still leads at 88.8. Pricing is wild at $2 per million input tokens and $6 per million output. The big news: open weights drop next week, including the 2.4T model and a 27B version that can run on a laptop. Anthropic and OpenAI might want to start worrying.
Orijinal gönderiyi gör@estebankiwi tarafından fotoğraf· Aug 3, 2026· Qwen
This is a bar graph showing performance metrics. The focus is on the bars themselves, representing different agents or models. The graph displays numerical scores above each bar, with the highest score of 88.8 belonging to the bar with the OpenAI logo. The style is clean and data-driven, with a prominent red outline around the first bar, which has a score of 86.6 and a blue recycling-like logo. The text "TerminalBench-2.1" and "Terminal Agent" are visible at the top.
