LLM & Chatbots
42 articles
LLM & Chatbots
Mistral's 1T 'Le Chonk' Sells a Refusal Gap, Not a Leaderboard Win
Mistral Large 4 is in preview with 49 billion active parameters, and the pitch is security work that closed models decline to do
LLM & Chatbots
A Cross-Chat Bug Let OpenAI's Model Decode Its Own Reasoning
OpenAI says operators pasted encrypted reasoning blobs from one conversation into another and asked the model to transcribe them. The flaw is patched; the company says the same path exists elsewhere.
LLM & Chatbots
OpenAI's Dots Agents Start at $100 a Month — and Pro 200 Loses Half Its Allowance
The always-on agents run on GPT-6 Astra with their own cloud computers, but the same DevDay that launched them also added a $500 tier and cut usage on the $200 plan.
LLM & Chatbots
GPT-6.1 Sol Holds Sol Pricing, Halves Cached Input to $0.10 per Million
OpenAI's DevDay model costs one-fifth of GPT-6 Astra's standard token rates while landing within a few points of it on the company's own agent benchmarks.
LLM & Chatbots
Anthropic's Sonnet 5.5 Beats Opus 5.5 on Terminal-Bench 4.0
The mid-tier model holds Sonnet 5's $2 per million input token price while Anthropic claims 30% more speed and up to 30% less cost per task
LLM & Chatbots
Opus 5.5 at Max Effort Waited 682 Seconds Before Its First Token
Artificial Analysis puts Anthropic's flagship configuration at 58 on its Intelligence Index against a median of 26 — reached with 260M output tokens and $5.98 per task.
LLM & Chatbots
The Best Model on OpenAI's New Mental Health Benchmark Scores 57.3%
More than 80 clinicians wrote the rubric. No model cleared 60% against it.
LLM & Chatbots
Ember-1 Costs the Same Per Token as Kimi K3 and Uses 40% Fewer of Them
Fireworks Research says training the model to shorten its reasoning traces — rather than turning the reasoning-effort dial down — is what preserved accuracy across seven benchmarks and two customers' production coding traffic.
LLM & Chatbots
Microsoft Quits the Consumer Chatbot Race and Puts Copilot on a Usage Meter
The rebuilt app folds the consumer assistant into work, adds Code and Autopilot tabs, and bills agentic features by consumption
LLM & Chatbots
Gemini's Live Avatar Holds Its Lip-Sync While Switching Among 97 Languages
Google made the animated persona generally available in Gemini Enterprise, with SynthID watermarks on every frame and custom faces locked behind an allowlist.
LLM & Chatbots
Claude Opus 5.5 Arrives 20% Cheaper — and With a Safeguard That Swaps In Older Models
The cache-read cut matters more than the headline price, and a benchmark footnote reveals when your request stops being served by Opus 5.5 at all
LLM & Chatbots
Grok 4.7 Holds Its Price and Doubles Its Terminal Score. The Bill Is Still the Argument.
SpaceXAI shipped a new frontier model a month after the last one without moving $2/$6 pricing — which reframes every benchmark gap as a cost-per-point question
LLM & Chatbots
OpenAI's Astra for Law Is Not a New Model. It Is a 230-Million-URL Search Index
A GPT-6 Astra configuration with a dedicated legal corpus lifted Vals AI Legal Research Bench correctness from 38.7% to 54%, and arrives with 26 partner plugins and an Am Law 200 access program
LLM & Chatbots
ChatGPT Picks a Favourite in 79% of Shopping Answers. Gemini Does It 7% of the Time.
An audit of 1,536 responses finds the two assistants barely cite the same sources — and that API-based research misses what shoppers actually see
LLM & Chatbots
AI Deck Generators Kept Dying at the Export Button. Claude Slides Ships PPTX.
Presentation software lives or dies on whether a colleague can open the file — the new document and slide editors hand back PowerPoint and PDF instead of a viewer link
LLM & Chatbots
iOS 27 Code Shows Siri's Server Model Can Be Swapped for Claude or GPT-5.6
Two private frameworks let outside models act as Siri extensions — or replace Apple's planner entirely
LLM & Chatbots
Mercury 2.5 Ships at 1,107 Tokens a Second and Four Cents a Million
Inception Labs is selling throughput, not intelligence — and customers in voice and coding agents are buying it
LLM & Chatbots
Claude Is Adults-Only, and Its Age Detector Keeps Locking Out Adults
Anthropic now publishes the mechanics of its 18+ enforcement, including a Yoti appeal route for users it wrongly suspends
LLM & Chatbots
Gemini Lands on Windows With an Alt+Space Shortcut and a Bid for Your Desktop
Google's native app puts its assistant one keystroke above whatever you are working on, globally, on Windows 10 and 11
LLM & Chatbots
ChatGPT Images 2.5 Halves Generation Latency and Adds a Sketch Canvas
Two new API models, in-image comments and templates push OpenAI's image stack toward a design tool.
LLM & Chatbots
Meta's Muse Agent Wants Your Inbox, Your Calendar and Your Credit Card
A free tier, $20 and $100 plans, and a virtual machine Meta says never sees your passwords.
LLM & Chatbots
OpenAI Ships a Data Agent and a Wall Street ChatGPT on the Same Day
One horizontal, one vertical — and together they show where enterprise revenue is pulling the company
LLM & Chatbots
Anthropic Warns Infostealer Malware Is Draining Paid Claude Accounts
Stolen browser sessions bypass MFA entirely, and Claude's usage dashboard cannot show subscribers who is spending their tokens
LLM & Chatbots
Silent Weight Swaps Are Testing What an API Model Name Guarantees
From September 14 the deepseek-v4-pro endpoint answers with V4.1 Flash weights — no opt-out, no parallel window, no replacement Pro announced
Mistral's 1T 'Le Chonk' Sells a Refusal Gap, Not a Leaderboard Win
Mistral Large 4 is in preview with 49 billion active parameters, and the pitch is security work that closed models decline to do
A Cross-Chat Bug Let OpenAI's Model Decode Its Own Reasoning
OpenAI says operators pasted encrypted reasoning blobs from one conversation into another and asked the model to transcribe them. The flaw is patched; the company says the same path exists elsewhere.
OpenAI's Dots Agents Start at $100 a Month — and Pro 200 Loses Half Its Allowance
The always-on agents run on GPT-6 Astra with their own cloud computers, but the same DevDay that launched them also added a $500 tier and cut usage on the $200 plan.
GPT-6.1 Sol Holds Sol Pricing, Halves Cached Input to $0.10 per Million
OpenAI's DevDay model costs one-fifth of GPT-6 Astra's standard token rates while landing within a few points of it on the company's own agent benchmarks.
Anthropic's Sonnet 5.5 Beats Opus 5.5 on Terminal-Bench 4.0
The mid-tier model holds Sonnet 5's $2 per million input token price while Anthropic claims 30% more speed and up to 30% less cost per task
Opus 5.5 at Max Effort Waited 682 Seconds Before Its First Token
Artificial Analysis puts Anthropic's flagship configuration at 58 on its Intelligence Index against a median of 26 — reached with 260M output tokens and $5.98 per task.
The Best Model on OpenAI's New Mental Health Benchmark Scores 57.3%
More than 80 clinicians wrote the rubric. No model cleared 60% against it.
Ember-1 Costs the Same Per Token as Kimi K3 and Uses 40% Fewer of Them
Fireworks Research says training the model to shorten its reasoning traces — rather than turning the reasoning-effort dial down — is what preserved accuracy across seven benchmarks and two customers' production coding traffic.
Microsoft Quits the Consumer Chatbot Race and Puts Copilot on a Usage Meter
The rebuilt app folds the consumer assistant into work, adds Code and Autopilot tabs, and bills agentic features by consumption
Gemini's Live Avatar Holds Its Lip-Sync While Switching Among 97 Languages
Google made the animated persona generally available in Gemini Enterprise, with SynthID watermarks on every frame and custom faces locked behind an allowlist.
Claude Opus 5.5 Arrives 20% Cheaper — and With a Safeguard That Swaps In Older Models
The cache-read cut matters more than the headline price, and a benchmark footnote reveals when your request stops being served by Opus 5.5 at all
Grok 4.7 Holds Its Price and Doubles Its Terminal Score. The Bill Is Still the Argument.
SpaceXAI shipped a new frontier model a month after the last one without moving $2/$6 pricing — which reframes every benchmark gap as a cost-per-point question
OpenAI's Astra for Law Is Not a New Model. It Is a 230-Million-URL Search Index
A GPT-6 Astra configuration with a dedicated legal corpus lifted Vals AI Legal Research Bench correctness from 38.7% to 54%, and arrives with 26 partner plugins and an Am Law 200 access program
ChatGPT Picks a Favourite in 79% of Shopping Answers. Gemini Does It 7% of the Time.
An audit of 1,536 responses finds the two assistants barely cite the same sources — and that API-based research misses what shoppers actually see
AI Deck Generators Kept Dying at the Export Button. Claude Slides Ships PPTX.
Presentation software lives or dies on whether a colleague can open the file — the new document and slide editors hand back PowerPoint and PDF instead of a viewer link
iOS 27 Code Shows Siri's Server Model Can Be Swapped for Claude or GPT-5.6
Two private frameworks let outside models act as Siri extensions — or replace Apple's planner entirely
Mercury 2.5 Ships at 1,107 Tokens a Second and Four Cents a Million
Inception Labs is selling throughput, not intelligence — and customers in voice and coding agents are buying it
Claude Is Adults-Only, and Its Age Detector Keeps Locking Out Adults
Anthropic now publishes the mechanics of its 18+ enforcement, including a Yoti appeal route for users it wrongly suspends
Gemini Lands on Windows With an Alt+Space Shortcut and a Bid for Your Desktop
Google's native app puts its assistant one keystroke above whatever you are working on, globally, on Windows 10 and 11
ChatGPT Images 2.5 Halves Generation Latency and Adds a Sketch Canvas
Two new API models, in-image comments and templates push OpenAI's image stack toward a design tool.
Meta's Muse Agent Wants Your Inbox, Your Calendar and Your Credit Card
A free tier, $20 and $100 plans, and a virtual machine Meta says never sees your passwords.
OpenAI Ships a Data Agent and a Wall Street ChatGPT on the Same Day
One horizontal, one vertical — and together they show where enterprise revenue is pulling the company
Anthropic Warns Infostealer Malware Is Draining Paid Claude Accounts
Stolen browser sessions bypass MFA entirely, and Claude's usage dashboard cannot show subscribers who is spending their tokens
Silent Weight Swaps Are Testing What an API Model Name Guarantees
From September 14 the deepseek-v4-pro endpoint answers with V4.1 Flash weights — no opt-out, no parallel window, no replacement Pro announced
Mistral's 1T 'Le Chonk' Sells a Refusal Gap, Not a Leaderboard Win
A Cross-Chat Bug Let OpenAI's Model Decode Its Own Reasoning
OpenAI's Dots Agents Start at $100 a Month — and Pro 200 Loses Half Its Allowance
GPT-6.1 Sol Holds Sol Pricing, Halves Cached Input to $0.10 per Million
Anthropic's Sonnet 5.5 Beats Opus 5.5 on Terminal-Bench 4.0
Opus 5.5 at Max Effort Waited 682 Seconds Before Its First Token
The Best Model on OpenAI's New Mental Health Benchmark Scores 57.3%
Ember-1 Costs the Same Per Token as Kimi K3 and Uses 40% Fewer of Them
Microsoft Quits the Consumer Chatbot Race and Puts Copilot on a Usage Meter
Gemini's Live Avatar Holds Its Lip-Sync While Switching Among 97 Languages
Claude Opus 5.5 Arrives 20% Cheaper — and With a Safeguard That Swaps In Older Models
Grok 4.7 Holds Its Price and Doubles Its Terminal Score. The Bill Is Still the Argument.
OpenAI's Astra for Law Is Not a New Model. It Is a 230-Million-URL Search Index
ChatGPT Picks a Favourite in 79% of Shopping Answers. Gemini Does It 7% of the Time.
AI Deck Generators Kept Dying at the Export Button. Claude Slides Ships PPTX.
iOS 27 Code Shows Siri's Server Model Can Be Swapped for Claude or GPT-5.6
Mercury 2.5 Ships at 1,107 Tokens a Second and Four Cents a Million
Claude Is Adults-Only, and Its Age Detector Keeps Locking Out Adults
Gemini Lands on Windows With an Alt+Space Shortcut and a Bid for Your Desktop
ChatGPT Images 2.5 Halves Generation Latency and Adds a Sketch Canvas
Meta's Muse Agent Wants Your Inbox, Your Calendar and Your Credit Card
OpenAI Ships a Data Agent and a Wall Street ChatGPT on the Same Day
Anthropic Warns Infostealer Malware Is Draining Paid Claude Accounts
Silent Weight Swaps Are Testing What an API Model Name Guarantees
LLM & Chatbots Most Read
Mistral's 1T 'Le Chonk' Sells a Refusal Gap, Not a Leaderboard Win
A Cross-Chat Bug Let OpenAI's Model Decode Its Own Reasoning
OpenAI's Dots Agents Start at $100 a Month — and Pro 200 Loses Half Its Allowance
GPT-6.1 Sol Holds Sol Pricing, Halves Cached Input to $0.10 per Million
Anthropic's Sonnet 5.5 Beats Opus 5.5 on Terminal-Bench 4.0