Google launches Gemini Decision 1 for batch closed-ended scoring
The Decisions API scores up to 1,000 questions in one pass and returns probabilities in 150 to 250 milliseconds.
Stories per day · 30 days
7 RSS
Relatedgoogle5openai5gemini-4-argon4ai-agents2anthropic2benchmarks2claude2compaction2frontier-model2gemini2inference2model-distillation2model-release2multimodal2reasoning-traces2vani2algapsa1allowlisting1
87 stories, newest first · one card per story
Stories tagged #api
EventSourcingDB 1.3.0 adds a v2 API and faster queries
v1 stays until January 31, 2027, but a data directory opened by 1.3 cannot be opened by earlier versions.
Vani hosts Kimi K3.1 Pro with its refusal layer removed
The Moonshot model runs on Vani's own GPUs with 1 million tokens of context, priced at $5 input and $15 output per million tokens.
Shopify Mandates IP Allowlisting for All Public Apps
Public apps must enable IP allowlisting by November 12, 2026 or face delisting from the Shopify App Store after November 20.
Apple releases App Store Connect API 4.6
Apple provides the App Store Connect API 4.6 for developers to download and review release notes.
Google identifies LLMjacking attacks generating $1 million in cloud AI charges
Researchers observe LLMjacking attacks using stolen API keys leading to fraudulent charges exceeding $100000 per day at some organisations.
1
Anthropic's Message Threads cut time to first token by 35 percent for long conversations
Message Threads on Claude platform transmits only changed content since last message, reducing time to first token by 35 percent for long conversations.
Simon Willison sends October newsletter
Sponsor-only newsletter covers LLM releases, API credits, and other tech news.
Linux syscall tables for 27 architectures updated with JSON exports and API
The updated site covers syscalls from Linux 3.3 to 7.3 across 27 architectures and offers kernel selectors plus JSON downloads of syscall history.
Apache Fineract through 1.15.0 has SQL injection in ORDER BY (CVE-2026-65874)
Caller-supplied orderBy, sortOrder and related parameters are concatenated on Center and datatable read APIs.
Vani adds OpenAI GPT-6.1 Terra to chat and API
The API id is openai/gpt-6.1-terra, and Vani places the model below GPT-6.1 Sol for everyday agent work.
OpenAI compaction flaw let researchers extract hidden model reasoning
Xint hid GPT-5.6 Sol reasoning in a compaction and recovered it through GPT-4.1 and GPT-4o; OpenAI paid $6000.
1
OpenAI API compaction flaw leaked hidden model reasoning traces
Xint says OpenAI confirmed the bug, paid $6,000, and shipped an undisclosed fix on October 23.
1
Researchers bypass OpenAI’s August fix for reasoning-trace theft
Xint recovered raw traces from gpt-4.1 and gpt-4o via compaction and tab substitution; OpenAI confirmed, mitigated, and paid a bounty.
1
OpenAI introduces GPT-6.1 Terra for computer use
Standard API pricing is $1 per million input tokens and $5 per million output tokens, 50% below prior Terra and Sol rates.
Xerj Console API bypass lets token holders mint credentials
CVE-2026-97406 lets a valid API token spawn persistent successor credentials that survive parent-token revocation; no patch is available.
2
Extend launches Verify API to flag fraudulent documents for AI agents
Extend says the API caught 96.3% of fraudulent documents at a 4.0% false-positive rate.
1
Anthropic Console overcharged my card 100x due to internationalization input bug
A developer says an Anthropic Console locale bug charged $569.52 instead of $5.04 for API credits.
Elorian’s first AI model wants to reason with images, not just about them
Elorian released Glimpse, a research-preview model that revisits images mid-reasoning through a visual scratch space.
CenterPoint Energy Data Breach: More Than 1 Million Texans Affected
CenterPoint Energy told Texas a breach exposed personal data of 1,062,060 residents, including Social Security numbers.
1
Huntress + AlgaPSA Integration: Built in Days by Nine Minds
Nine Minds shipped an AlgaPSA integration that turns Huntress incident reports into tickets using the public API.
[AINews] AI Engineer NYC: Gemini 4 Argon, Vals AI Benchmarks
Google made Gemini 4 Argon generally available, with independent benchmarks placing it near rival frontier models.
11
Google Releases Gemini 4 Argon API With 1M Token Output Limit
Google opened Gemini 4 Argon on the paid API, a sparse MoE model with a combined 1M-token window.
11
llm-gemini 0.35
llm-gemini 0.35 adds Gemini 4 Argon, priced at $2 and $10 per million tokens.