Trump reportedly considering Scott Bessent for a new White House AI czar role
Citing Semafor, Polymarket says Trump is reportedly considering Treasury Secretary Scott Bessent for a new White House “AI czar” role.
AI News
Instacart says “Taco Tuesday” will soon start a grocery cart on Muse
Instacart is coming to Muse. It says people will soon be able to connect Instacart, turn “Taco Tuesday” into a cart from a favorite store, and check out for home delivery.
Claude Opus 5.5, GPT-6 Luna and GPT-6 Sol are available in Glean
Glean says Claude Opus 5.5 cuts token costs by 50% versus its predecessor on Glean's own evaluations, and that GPT-6 Luna and GPT-6 Sol each cost 50% less than their predecessors.
GPT 6 Sol and Luna are live in Command Code
Luna is available on all plans and the API; the $10 GOAT plan is given as an example with about 30,000 requests and about 1.5 billion tokens. Sol is available on the Pro and Max plans and the API.
Databricks launches OpenAI GPT-6 Sol, GPT-6 Luna, and AnthropicAI Claude Opus 5.5
Databricks is making OpenAI’s GPT-6 Sol and GPT-6 Luna and AnthropicAI’s Claude Opus 5.5 available to its customers. Databricks says the GPT-6 models perform strongly at lower cost with a higher Pareto frontier on OfficeQA Pro, and that Claude Opus 5.5 improves cost efficiency on enterprise document parsing.
Anna Tong reports micro1_ai and aliansarinik raised over $100 million at a $4 billion valuation
Anna Tong reports a raise of more than $100 million and says annual recurring revenue grew from $7 million to $500 million in 1.5 years at a business she describes as selling data to AI labs.
GPT-6 prompt caching can be tuned with breakpoints and prewarming
OpenAI Developers says applications can select reusable prompt prefixes with explicit cache breakpoints, change reasoning effort and tool availability while preserving cached context, and prewarm shared context so responses start sooner.
OpenAI Developers improves prompt caching for GPT-6 in the API
Higher default cache-hit rates let more GPT-6 input tokens receive cached-input discounts of up to 90%. OpenAI Developers says the change helps agents run faster and cost less.
OpenAI Developers introduces a Prompt Caching Dashboard and diagnostics API
A new Prompt Caching Dashboard on platform.openai.com tracks cache-hit rates. A diagnostics API identifies changes that prevented cache reuse and estimates how many tokens were affected.
Coinbase launches stock and ETF trading for AI agents, Polymarket says
Users can authorize the agents to research markets, pay for data, and execute trades, according to Polymarket.
ComfyUI says MiniMax H3 video VAE is about 2x faster on NVIDIA GPUs
In ComfyUI, encoding is up to about 2.2 times faster and decoding is about 1.4–2.7 times faster. ComfyUI reports a 1344×768, 129-frame round trip falling from 24.3 seconds to 12.7 seconds.
Grok 4.7 (xHigh) ranks 10th in Code Arena WebDev
Arena.ai scores SpaceXAI’s Grok 4.7 (xHigh) at 1,632 points, 16 points and six places above Grok 4.6 (High). Simulations and Consumer Product also move up.
Polymarket says Expedia announced a Meta Muse integration for AI travel planning
According to Polymarket, the integration will let AI agents help users plan travel, including finding and booking hotels.
Perplexity says a later Computer checkpoint reduced tool-call failures 21.2% in a live A/B test
Perplexity post-trained a Computer model to learn from its own errors using hint-guided self-distillation. It said the 21.2% reduction was relative to an earlier checkpoint.
Pika’s Swap Anything app can change a video’s cast, wardrobe, setting, or props
The app is now available on the new Pika, Pika’s AI creative platform. It keeps the original video’s timing, motion, and narrative while replacing the selected elements.
Sluicebox agent Lucy uses NVIDIA Nemotron 3 Ultra for data-center carbon data
Sluicebox’s AI supplier agent Lucy uses NVIDIA Nemotron 3 Ultra to collect and check supplier data while engineers design a data center’s carbon footprint. In Sluicebox’s tests, Nemotron improved accuracy, cut costs by 51–80%, and returned responses up to 2x faster.
Codex CLI 0.156.0 adds optional fullscreen TUI and default voice
The release also includes transcript search, mouse copy, an F8 voice toggle and /voice settings, plus a /usage dashboard for tokens, plugins, and skills.
Four labs lead open-weight TTS across nine language leaderboards
Artificial Analysis finds no single open-weight text-to-speech model ahead on all nine new language leaderboards. Leadership is split among Higgs Audio V3 TTS, OpenAudio S1 Mini, Voxtral TTS and Breeze TTS 2.
Artificial Analysis adds TTS leaderboards for nine languages
Artificial Analysis has extended the Controlled Voice Arena with separate text-to-speech leaderboards for Japanese, Mandarin Chinese, Hindi, Spanish, German, French, Portuguese, Vietnamese, and Arabic. Rankings use native-language prompts, standardized cloned voices, and preference votes from first-language speakers.
jlehman_ adds decision model support to OpenClaw core and plugins
OpenClaw says the support was pushed last week, and describes how it is thinking about using decision models and how people can contribute.