RuntimeWire

Real-time startup intelligence for the AI economy. Funding, model launches, infra shifts, and founder moves — the signals that matter.

Latest stories — page 3

• Head to head: grok-4.3 vs Llama-4-Scout-17B-16E-Instruct

  One model was consistently cleaner, sharper, and better at following instructions; the other kept leaving small but consequential mistakes on the table. Across coding, transformation, summarization, and business writing, this matchup wasn’t especially close.

• Blueshoe launches an AI-native law firm for consumer and plaintiff cases

  YC's Spring 2025 company pairs a Harvard Law CEO with an MIT engineer and says legal intake can start inside chat apps.

• Probook raises $40 million by putting AI at the dispatch board

  George Eliadis started with missed pressure-washing calls. Sequoia and a16z are betting that dispatch is the control point in home services.

• Samsung SDI backs Forge Nano's defense battery push

  Samsung SDI's money and engineering help would give Forge Nano manufacturing credibility as it courts defense buyers for U.S.-sourced cells.

• Palmer Luckey's Anduril is chasing a Nissan factory for Japan's drone buildout

  Reuters says Anduril is in talks for Nissan's Oppama plant, a civilian auto site that could become a local defense production hub.

• Chris Lattner's Modular is selling to Qualcomm in a bet on the AI runtime layer

  The deal puts Mojo, MAX and Modular's hardware-neutral pitch inside Qualcomm's edge-to-cloud silicon strategy.

• Intercept launches with $500M to attack colds, flu and indoor-air transmission

  Nan Ransohoff, Charlie Petty and Devin Sok are using grants, investments and buyer demand to pull respiratory prevention out of a funding gap.

• Sony opens U.S. pre-orders for its wearable cooling device

  The REON POCKET PRO Plus costs $259.99 in the U.S., with Sony estimating late-July delivery for the neck-worn thermal gadget.

• OpenAI Updates GPT-5.5 Instant to Better Handle Intent, Advice and Shopping

  The June 24 release targets ChatGPT's default workhorse model, with OpenAI pushing it deeper into recommendation and decision queries.

• Vishal Sikka raises $32M for Hang Ten, an AI bet against the IT services model

  Mayfield led the seed round for Hang Ten Systems, which says AI can compress the work behind enterprise software projects.

• Anthropic Says Alibaba-Linked Operators Used 25,000 Accounts to Mine Claude for Qwen

  The Claude maker says the April-to-June campaign produced 28.8 million exchanges and targeted coding and agentic reasoning.

• Head to head: Bernini-R Edit Image vs Happy Horse 1.1 Image to Video

  This matchup turns on execution, not vibes. Bernini-R Edit Image flashes style, but Happy Horse 1.1 Image to Video is the model that more consistently obeys the brief, preserves scene logic, and delivers cleaner motion storytelling across both tests.

• Coval raises $28M to test voice AI like Waymo tested cars

  Norwest led Brooke Hopkins' Series A as enterprises push voice agents from demos into production calls.

• Head to head: Bagel vs Krea 2 Large

  This matchup splits cleanly between Bagel’s eye for composition and Krea 2 Large’s stricter obedience to the brief. One model makes prettier images when it gets room to improvise; the other wins by actually delivering what was asked for.

• Anthropic's Fable 5 return signals are showing up in Claude Code and AWS

  The model remains offline after a US export-control order, but product strings, AWS docs, and a new lawsuit point to a negotiated relaunch path.

• Head to head: Anthropic: Claude Opus 4.8 vs Google: Gemini 3.5 Flash

  This one is close on the aggregate, but the split tells a clear story: Gemini 3.5 Flash wins by being more disciplined about format and slightly sharper on practical instruction-following. Claude Opus 4.8 lands the strongest single extraction/summarization performance, yet gives away too much on avoidable execution det

• Seltz Lands $12.5 Million to Build Search Infrastructure for AI Agents

  Antonio Mallia, an Amazon and Pinecone veteran, is betting agents need their own crawler, index, retrieval models, and ranking stack.

• Meta turns Facebook Creator Studio into an AI companion app

  The standalone app is being tested with select creators and folds in Facebook's AI assistant for analytics, comments and content planning.

• OpenAI's Jalapeno chip moves the AI fight from GPUs to unit economics

  The Broadcom-built inference ASIC is slated for deployment by year-end, but OpenAI has not released final benchmarks, pricing or yields.

• Greptile puts numbers on the AI pull request spam problem

  Rahul Bathija's OpenClaw study gives Daksh Gupta's code-review startup a live dataset for its validation-layer thesis.

• Ryan Cohen pulls his GameStop pay plan as the eBay bid becomes an incentive test

  The Chewy cofounder is trying to remove a clean objection from critics: that GameStop's takeover push could enrich him personally.

• Slate Auto opens EV truck preorders at $24,950, turning its reservation list into the real test

  The Troy, Michigan EV startup is asking buyers for a $300 non-refundable deposit before production starts in autumn 2026.

• Bloomberg: Dutch chip startup Euclyd is seeking about 200 million euros for a Series A, not yet closed

  Bloomberg reports Euclyd has several term sheets, but no named lead investor or valuation, and the amount could change.

• Racket's Rhombus 1.0 turns a decades-old macro thesis into a cleaner language bet

  Matthew Flatt, Robby Findler, Sam Tobin-Hochstadt and a 20-person contributor group are trying to bring Racket-style extensibility beyond Lisp syntax.