RuntimeWire

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

Latest stories — page 10

• MiniMax says it raised $2B, but leaves investors and valuation unnamed

  Ryan Lee's X thread did not name investors, valuation or deal structure, leaving the financing terms around the Chinese AI lab unresolved.

• Fleek raises $25M to turn secondhand clothing sorting into software

  Burda Principal Investments led the Series B, with eBay joining a cap table that already included a16z, HV Capital and Y Combinator.

• Stan's 14-day AI build turned customer knowledge into $3 million ARR

  John Hu says Stanley began with manual interviews, public posts and AI-written cold emails before expanding from LinkedIn to Instagram.

• GPT-5.6, OpenAI's flagship model, helps build itself

  Shruti Gandhi's July 9th post spotlights a deeper shift behind Sol, Terra and Luna: the frontier model is becoming production infrastructure.

• Trinity Robotics raises EUR500k to scale Konyk ground robots

  The Ukrainian maker of casualty-evacuation UGVs says Konyk ONE is already used by more than 20 military units across frontline sectors.

• The People Who Monetized Mobile Apps Are Building the Ad Layer for AI Chats

  Velocity raised a $27 million seed round to put native ads, auctions, and intent analysis inside AI applications.

• Skapion lands $36 million seed round for counter-swarm air defense

  UP Partners and Khosla Ventures co-led the seed round for the Washington and Ramat Gan defense startup.

• Europe Just Put Chips Act Money Behind Diamond-Based Chip Inspection

  Kevin Berghoff and Dr. Fleming Bruckmaier are scaling a Munich factory for diamond-based quantum microscopes used in semiconductor failure analysis.

• QIZ Security's $17M seed puts enterprise crypto inventory ahead of Q-Day

  Ben Volkow's startup announced a $17 million seed round today as regulators and CISOs move from PQC theory into migration work.

• Sherpa raises $2.2M to manage contractors, suppliers and AI agents in one workflow

  The German company is selling enterprises on a single operating layer for external work, from intake and scoping to compliance and payment.

• reverse.fashion adds HTGF backing to automate textile sorting

  The Berlin TU spin-off raised a seven-figure pre-seed extension as EU textile rules push sorters toward data-rich reuse.

• Head to head: Muse Spark 1.1 vs GLM 5.2

  This one wasn’t a blowout, but Muse Spark 1.1 did enough across the harder, failure-prone tasks to come out ahead. GLM 5.2 is the tidier formatter in spots, yet Muse’s edge on core correctness gives it the lean verdict.

• Anthropic Puts a Meter on Claude Fable 5

  Starting July 12, paid Claude subscribers will owe API-style usage fees for Anthropic's top consumer model, a test of whether flat-rate AI subscriptions can survive agentic compute costs.

• Alison Quinn's Respiro Diagnostics raises GBP 1M for breath-based lung diagnostics

  The UK company plans a September 2026 lung cancer clinical study after early proof-of-concept work in mesothelioma patients.

• Aardaia raises 5 million euros to domesticate a protein tuber for European fields

  Point Nine led the seed round for the Wageningen startup, which is turning a wild crop into a non-GMO alternative to imported protein.

• Cells2Pixels gives neural cellular automata an 8x resolution path

  Ehsan Pajouheshgar's SIGGRAPH 2026 project keeps the automata grid coarse and uses a small local decoder to render detail.

• Head to head: AuraFlow vs Seedream 5.0 Pro Image Editing

  One model controlled the brief; the other mostly looked for ways to drift off it. Across eight image-editing tests, Seedream 5.0 Pro Image Editing turned a lopsided scoreline into a no-argument verdict.

• The AO is digitising battlefield evacuations after its founder learned the problem in Ukraine

  Howard Hunt built The AO in Berlin around a hard lesson from Donetsk Oblast: drones have turned casualty evacuation into a data problem.

• GPT-5.6 Built a Tiny iMessage Transformer From One Prompt

  Pietro Schirano says Sol generated the code, pipeline, tokenizer, training loop and local reply model in about 30 minutes.

• Google trains SensorFM on a trillion minutes of wearable data

  The July 9th release turns Fitbit and Pixel Watch signals into a reusable health model evaluated across 35 prediction tasks.

• Mercor seeks $20B valuation as AI training spend pours into human data

  The AI training startup also said it will acquire Deeptune, pushing beyond expert matching into the environments used to train agents.

• Head to head: Muse Spark 1.1 vs Kimi-K2.7-Code

  This one is effectively even. Muse Spark 1.1 and Kimi-K2.7-Code trade narrow wins on formatting, coding, and language tasks, and the aggregate gap is small enough that calling a real leader would be overstating the evidence.

• GPT-5.6 Sol runs a week-long voxel Manhattan build in Matt Shumer demo

  The HyperWrite co-founder says OpenAI's flagship preview model completed the job autonomously, with no prompt, repo, logs, or cost disclosed.

• Aleph's silent-speech prototype reads tongue motion at 15.6% word error rate

  The research lab says a 50-hour dataset let its prototype decode open-vocabulary speech, with hardware still stuck at probe-and-gel stage.