Google releases Gemini Decisions API and Gemini Decision 1 model for fast multimodal closed-ended decisions
Google has launched the Gemini Decisions API, powered by the new Gemini Decision 1 model, which can evaluate up to 1,000 yes/no, multiple-choice and rating questions in a single pass across text, images, audio, video and PDFs (up to 256K tokens) with 150-250ms end-to-end latency.
模型发布/更新
-
Baidu releases ERNIE 5.5 Flash Preview, weights to be open-sourced by November 25
Baidu has released ERNIE 5.5 Flash Preview, the first post-trained version based on the earlier Base model, with 280B…
-
Runway releases open-weight 27B world action model Praxis-1 under Apache 2.0
Runway says its world action model Praxis-1 is now available to all as open weights under Apache 2.0. The model is a…
-
Google DeepMind 发布 AlphaGenome 2,预测基因变异对特定细胞类型的影响
Google DeepMind 发布下一代 DNA 序列模型 AlphaGenome 2,可预测 DNA 变异如何影响数百种人类细胞类型中的基因表达。模型改为条件式架构,训练数据超过 1.07 亿个单细胞样本、覆盖 68 种组织,在 Te…
产品发布/更新
-
Google releases Gemini Decisions API and Gemini Decision 1 model, focused on fast multimodal closed-ended decisions
Google launched the Gemini Decisions API, powered by the new Gemini Decision 1 model, which can evaluate up to 1,000…
行业动态
-
OpenAI Q3 revenue reaches $10 billion, efficiency still lags Anthropic
According to reports from The Wall Street Journal and The Information, OpenAI's Q3 revenue grew about 50%…
-
Grok accused of generating child sexual abuse imagery as xAI faces multiple class actions
According to Bloomberg, multiple minors have filed class-action lawsuits against Musk's xAI over Grok being used to…
-
Security firm Claroc: 94.7% of Grok-built apps accessible by anyone, including patient data and SpaceXSI internal files
Security firm Claroc reports that of 37,571 Grok-built apps still online that it examined, 35,578 (94.7%) could be…
-
Anthropic's amended S1 filing discloses new compute deals with Microsoft and Google exceeding $10 billion
Anthropic disclosed in its amended S1 filing that it has reached new agreements with Microsoft and Google exceeding…
论文研究
-
Anthropic releases MAMUBench: single coordinating agent outperforms per-user agent teams across four environments
An Anthropic Fellows Program team built four multi-user multi-agent simulated environments (shared API token budget,…
技巧与观点
-
OpenRouter tests context token costs of MCP vs CLI vs Skills
OpenRouter tested the token costs of MCP, CLI and skills on the same GitHub task: on Claude Sonnet 5.5, CLI used 1.
-
OpenRouter tests impact of fp8 and fp4 quantization on model quality and price
OpenRouter published an explainer on quantization, covering the principles of LLM quantization and reporting tests…
-
Claude Fable 5.2 hands-on: large models are devouring aesthetics and creation through code
After testing Claude Fable 5.2 against Gemini 4 Carbon, the author argues that Fable 5.2's standout difference lies in…