Every event in the Vikshy Score ledger, newest first, each with its original source, the exact line we relied on, and how it feeds a provider's momentum. 75 events as of 2026-09-02 06:00 UTC.
Each entry shows its factor (which of the seven scored dimensions it moves), its evidence tier (how directly we can verify it), and a trail: direction · base points(materiality, tier) · decay(age / half-life) = contribution now. Contributions are per-event and pre-cap; the engine then applies cluster, daily-flow and community caps before a factor's momentum tilts the published score. Full formula on the methodology page.
Cohere announced North Automations, a new intelligent workflow orchestration product.
Introducing North Automations: Intelligent workflow orchestration
Google announces Gemini and Pixel partnerships with five global football clubs to enhance fan matchday experiences.
Google Gemini and Pixel partner with five global football clubs to elevate the fan matchday experience through AI and Smartphone Technology.
Cohere announced the introduction of North Automations, an intelligent workflow orchestration product.
Introducing North Automations: Intelligent workflow orchestration
Google launched Sheets canvas, a feature that turns spreadsheet data into interactive dashboards and trackers via prompts.
Sheets canvas turns data into interactive dashboards, custom study trackers, seating charts, and more, all with a simple prompt.
OpenAI previews Ultrafast, a new API service tier that runs GPT-5.6 Sol up to 14x faster with 750 tokens per second.
Preview Ultrafast, a new OpenAI API service tier that runs GPT-5.6 Sol up to 14× faster. Powered by Cerebras, it delivers up to 750 output tokens per second.
RingCentral is highlighted as using ChatGPT Work and Codex to accelerate AI product development.
See how RingCentral uses ChatGPT Work and Codex to accelerate AI product development and centralize operational intelligence across engineering and operations.
Cohere launched North Automations, an intelligent workflow orchestration product.
Introducing North Automations: Intelligent workflow orchestration
Google announces AMIE, a research medical AI system, demonstrating real-time clinical video consultation capabilities in simulated settings.
Google introduces AMIE for real-time clinical video consultations in simulated settings.
OpenAI announces testing ads in ChatGPT to support free access.
OpenAI begins testing ads in ChatGPT to support free access
OpenAI's Daybreak cybersecurity models are now available on AWS through Amazon Bedrock.
OpenAI and AWS are making Daybreak cybersecurity capabilities available through Amazon Bedrock to support enterprise security workflows.
Google announces new AI and agentic experiences across Google Ads and Analytics to simplify marketing workflows.
Learn how new AI and agentic experiences across Google Ads and Google Analytics can simplify your marketing workflow.
OpenAI reports that Model ML uses GPT-5.6 Sol for finance work, indicating adoption in financial services.
Model ML uses GPT-5.6 Sol to carry finance work from research and analysis through editable, traceable PowerPoint decks and Excel workbooks.
OpenAI released GPT-5.6-Cyber, a cybersecurity-specific model available through Daybreak Red.
Meet GPT-5.6-Cyber, OpenAI’s cybersecurity-specific model available through Daybreak Red for authorized vulnerability research, exploit validation, and security testing.
Zapier's enterprise marketing team uses ChatGPT Work to optimize lead funnel, create campaign assets, and automate reporting.
The enterprise marketing team at Zapier uses ChatGPT Work to reduce the number of drop-offs in its lead funnel, build campaign assets, and automate reporting.
Virgin Atlantic adopts ChatGPT Work for customer journey analysis and decision-making.
Virgin Atlantic is accelerating research, product planning, and decision-making with ChatGPT Work, helping teams connect signals across the customer journey.
OpenAI published preliminary cybersecurity evaluations for Astra and announced enhanced security controls.
OpenAI is sharing preliminary cybersecurity evaluations for Astra and the steps we’re taking to strengthen safeguards and security controls.
HSP GRUPPE adopts ChatGPT Enterprise to enhance productivity and service capacity in tax advisory.
Discover how HSP GRUPPE uses ChatGPT Enterprise to boost productivity, improve work quality, and create more capacity for tax advisory and client service.
OpenAI improves GPT-5.6 Sol in ChatGPT and expands access to GPT-5.6 Luna for free users.
ChatGPT introduces improved GPT-5.6 Sol with better accuracy and consistency, plus expanded access for free users and unlimited everyday chats with GPT-5.6 Luna.
OpenAI partners with the American Psychological Association to advance responsible AI guidance and safeguards for youth mental health.
OpenAI and the American Psychological Association advance evidence-based guidance, resources, and safeguards for responsible AI use and youth mental health.
Google and Kaggle completed a free AI Agents Intensive course with 353,000 participants.
Kaggle’s AI Agents Intensive with Google brought learners together in a no-cost course to build and deploy the next frontier of AI.
OpenAI customer Circles reports increased ARPU by 22% and reduced churn by 9% using OpenAI API and Codex for telco personalization.
Circles uses the OpenAI API and Codex to power AI-native telco experiences, increasing ARPU by 22%, reducing churn by 9%, and improving development efficiency.
Univé is adopting ChatGPT Enterprise to build an AI-ready workforce with leadership and governance.
See how Univé built an AI-ready workforce with ChatGPT Enterprise by combining leadership, responsible governance, and employee-led innovation to transform work at scale.
OpenAI announces lower pricing for GPT-5.6 models on Luna and Terra.
Explore lower GPT‑5.6 pricing for Luna and Terra, and how OpenAI’s more efficient models help enterprises deploy AI workflows at scale.
Avatarin deployed OpenAI's GPT-Realtime for 24/7 retail support at Yamada Denki, reaching 30,000 users with 92% positive feedback.
In two weeks, 30,000 people used the agent and 92% of survey responses were positive.
Enabling two API settings tripled GPT-5.6 ARC-AGI-3 scores.
How two API settings improved GPT-5.6 performance on ARC-AGI-3, boosting scores and efficiency by retaining reasoning and enabling compaction.
OpenAI grants 100,000 academic researchers free access to advanced ChatGPT models.
OpenAI is giving 100,000 academic researchers free access to ChatGPT's most advanced AI models to accelerate scientific research, collaboration, and discovery.
OpenAI launches GPT-5.6, improving efficiency across models, inference, and agentic workflows.
GPT-5.6 improves AI efficiency across models, inference, and agentic workflows, helping deliver more useful intelligence per dollar.
Google announces new capabilities in Managed Agents for the Gemini API.
We’re announcing even more new capabilities in Managed Agents in Gemini API so developers can build reliable, production-ready agents.
Anthropic introduces Claude Opus 5, a step change improvement for the Opus tier powering long-running agents with improvements in coding and professional work.
Opus 5 is a step change improvement for the Opus tier powering long-running agents while delivering improvements in coding and professional work.
OpenAI launches Health in ChatGPT for eligible U.S. users, enabling secure connection of medical records and Apple Health data.
Health in ChatGPT now lets eligible U.S. users securely connect medical records and Apple Health to get more personalized insights and better understand their health.
NTT DATA Group adopts ChatGPT Enterprise and Codex, reducing incident analysis to 30 minutes for 9,000 employees.
NTT DATA Group uses ChatGPT Enterprise and Codex to help 9,000 employees automate work, cut incident analysis to 30 minutes, and scale secure AI adoption.
OpenAI launches a program for small businesses to adopt ChatGPT Work.
OpenAI launches the ChatGPT for Small Businesses program, helping entrepreneurs build AI skills, automate work, and grow with ChatGPT Work.
Meta's AI models are being used in the first wave of Genesis Mission projects at Lawrence Berkeley National Laboratory.
How Meta’s AI Models Are Powering the First Wave of Genesis Mission Projects
Google launches Gemini Omni and personal avatars in Google Vids, enhancing video creation capabilities.
Gemini Omni and personal avatars in Google Vids make video creation easier than ever.
On July 15, 2026, China's cyberspace regulator approved Apple's AI services for the China market, built on a deal to power Apple Intelligence experiences with Alibaba's Qwen model, a major distribution win.
Apple Intelligence approved for launch in China with Alibaba's Qwen AI.
DeepSeek's V4 line (V4-Pro 1.6T MoE and V4-Flash 284B, 1M-token context) moved from its April preview to an official mid-July 2026 release with stated performance and feature optimizations.
DeepSeek to launch V4 in mid-July with new peak-time API pricing.
On the morning of July 15, 2026, ChatGPT suffered a global outage affecting login, chat and image generation via upstream connection errors; OpenAI's enterprise API and Codex stayed operational and service was restored the same day.
ChatGPT is back online after a major global outage that left thousands of users unable to access the service.
On July 14, 2026, Anthropic launched Claude for Teachers, giving verified US K-12 educators a free year of premium Claude access with standards-aligned curricula for all 50 states and connectors to K-12 tools.
Claude for Teachers gives verified K-12 educators in the US free access to premium Claude capabilities, mapped to academic standards in all 50 states.
Google missed its promised June launch for flagship Gemini 3.5 Pro, scrapping the original base model and restarting pretraining after engineers found structural failures in recursive tool-calling and SVG generation; the target slipped to July 17 with no official model card or pricing.
Google scrapped the original Gemini 3.5 Pro base model and restarted pretraining, pushing the launch to July 17, 2026.
On Artificial Analysis, GPT-5.6 Sol (max) scored 59 on the Intelligence Index, one point behind Claude Fable 5, and set a new state of the art of 80 on the Coding Agent Index while using far fewer tokens and less time.
GPT-5.6 Sol with max reasoning sets a new state of the art at 80 on the Coding Agent Index, 2.8 points above Fable 5, using less than half the output tokens.
OpenAI released its new flagship GPT-5.6 model family on July 9, 2026, comprising Sol (flagship), Terra (mid-tier) and Luna (fastest/cheapest), with broad availability across ChatGPT and the API.
GPT-5.6: Frontier intelligence that scales with your ambition.
GPT-5.6 launched with per-million-token API pricing of Sol $5 input / $30 output, Terra $2.50 / $15, and Luna $1 / $6, positioning Sol at roughly a third of Claude Fable 5's cost per task.
Sol is $5 input / $30 output, Terra is $2.50 input / $15 output, and Luna is $1 input / $6 output per million tokens.
Alongside GPT-5.6, OpenAI launched ChatGPT Work on July 9, 2026, an agent that autonomously completes multi-step tasks and delivers finished outputs (docs, decks, spreadsheets, web apps) by connecting to apps like Gmail, Slack and Google Drive.
OpenAI releases GPT-5.6 and ChatGPT Work tool.
Anthropic rolled out Reflect, a built-in dashboard that lets users track and visualize how they use Claude and their broader AI habits.
Anthropic introduced Reflect, a built-in dashboard that lets users track and visualize how they use Claude.
Independent evaluator Artificial Analysis scored Grok 4.5 at 54 on its Intelligence Index (4th overall) and placed it 4th on GDPval-AA v2 with an Elo of 1543, between Opus 4.8 (1600) and GLM-5.2 (1513).
Artificial Analysis independently places it at 54 on its Intelligence Index, ranked #4 overall.
On July 8, 2026, OpenAI released GPT-Live-1 and GPT-Live-1 mini, conversational voice models it says sound more natural and handle turn-taking better than prior voice systems.
OpenAI released new conversational models called GPT-Live-1 and GPT-Live-1 mini, claiming that they sound more natural and can handle turn-taking better.
On July 8, 2026, SpaceXAI (xAI) released Grok 4.5, a Cursor-trained model focused on coding, agentic tasks, and knowledge work, available via the console, Grok Build, and Cursor.
Grok 4.5, SpaceXAI's smartest model built to excel at coding, agentic tasks, and knowledge work.
Grok 4.5 shipped at $2 per million input tokens and $6 per million output tokens ($0.50 cached input), pricing well below comparable frontier models such as GPT-5.5 and Opus-class rivals.
Grok 4.5 is priced at $2 per million input tokens and $6 per million output tokens.
SpaceXAI partnered with Cursor to train Grok 4.5 and made it available at launch in Cursor's desktop, web, iOS, and CLI interfaces across all plans.
Grok 4.5 is available today in Grok Build, in Cursor on all plans ... The model was trained alongside Cursor.
On July 8, 2026, Mistral released Robostral Navigate, an 8B embodied vision-language navigation model that steers robots using a single RGB camera and plain-language instructions.
Robostral Navigate ... an 8B vision-language model that lets robots follow natural language instructions ... with nothing more than a single ordinary RGB camera.
Mistral reported Robostral Navigate reaches 76.6% success on the R2R-CE benchmark for unseen environments, 9.7 points above the best single-camera approach and 4.5 points above leading multi-sensor systems.
achieves a 76.6% success rate on the R2R-CE benchmark for previously unseen environments, surpassing the best single-camera approach by 9.7 points.
Google announces new capabilities for Managed Agents in Gemini API, including background tasks and remote MCP support.
We’re announcing new capabilities in Managed Agents in Gemini API so developers can build reliable, production-ready agents.
In early July 2026, CEO Arthur Mensch confirmed a new open-weight Mixture-of-Experts model family, described as 'fat but sparse,' entering July partner early access as Mistral's next bid to close the frontier gap.
a new Mixture-of-Experts family described as 'fat but sparse' ... entering July 2026 early access.
On July 2, 2026, Second Front and Cohere announced deploying Cohere North in a live UAE edge environment aboard an Armada Galleon in about 90 minutes, demonstrating rapid sovereign-AI standup.
Second Front and Cohere Successfully Deploy Sovereign AI Infrastructure in UAE.
After redeployment, Claude Fable 5 again topped the Artificial Analysis Intelligence Index in July 2026 leaderboard readings, leading at ~59.9 ahead of GPT-5.6 Sol, with Anthropic models occupying the top ranks.
Claude Fable 5 leads the Artificial Analysis Intelligence Index leaderboard with 59.9%, ahead of GPT-5.6 Sol (58.9%).
On July 1, 2026, Anthropic restored Claude Fable 5 worldwide across the Claude Platform, Claude.ai, Claude Code and Cowork after a 19-day suspension, shipping a new safety classifier that blocks the flagged jailbreak technique in over 99% of cases.
The new classifier blocks the specific technique described in the report in over 99% of cases.
Reported July 1, 2026, Meta is building a cloud business, Meta Compute, to sell access to AI compute and its Muse models, competing with AWS, Azure, and Google Cloud; the news lifted Meta shares ~9%.
Meta pops 9% as company makes cloud push to sell excess AI compute power capacity.
LMArena leaderboard data placed Alibaba's Qwen 3.7-Max as the top-ranked Chinese model (around #5 globally in text), leading domestic rivals while still trailing top US models from Anthropic, Google, and OpenAI.
Alibaba teases new Qwen previews, highest-ranking Chinese AI models on Arena.
The US government export controls that had immediately suspended access to Claude Fable 5 and Mythos 5 on June 12 were lifted as of June 30, 2026, restoring regulatory clearance for Anthropic's frontier Mythos-class models.
As of June 30, the export controls on Fable 5 and Mythos 5 were lifted.
On June 30, 2026, Anthropic launched Claude Sonnet 5, its most agentic Sonnet yet, claiming near-Opus-4.8 performance at Sonnet pricing (72.7% SWE-bench Verified) and making it the default model for Free and Pro plans.
Claude Sonnet 5 delivers near-Opus-4.8 performance at Sonnet pricing, with particular strength in agentic, multi-step tasks; it scores 72.7% on SWE-bench Verified.
Claude Sonnet 5 launched at introductory pricing of $2 per million input tokens and $10 per million output tokens through August 31, 2026, rising to $3 / $15 afterward; a new tokenizer can map the same input to 1.0-1.35x more tokens.
Through August 31, 2026, Sonnet 5 runs at introductory pricing of $2 per million input tokens and $10 per million output tokens; after that it is $3 and $15.
On June 30, 2026, Google released Nano Banana 2 Lite, its fastest and most cost-efficient image model, generating a text-to-image in about four seconds via the Gemini API, AI Studio and the Gemini Enterprise Agent Platform.
Nano Banana 2 Lite is now available as the fastest and most cost-efficient Gemini Image model yet.
Google launched Gemini Omni Flash in public preview on June 30, 2026, a natively multimodal model that turns images into short videos and supports conversational, plain-language video editing.
Gemini Omni Flash, new in public preview, turns those images into video, edited in plain language.
Google positioned its new generative-media models on price, with Nano Banana 2 Lite at about $0.034 per 1K-resolution image and Gemini Omni Flash video at $0.10 per second.
Nano Banana 2 Lite makes a text-to-image in about four seconds at $0.034 a shot; Omni Flash turns images into video at $0.10 per second.
Announced June 30, 2026, DeepSeek's V4 official release introduces first-ever peak-hour surcharging: API prices double during Beijing 9-12 and 14-18 windows while off-peak prices hold.
prices double during 9AM-12PM and 2PM-6PM daily ... the first time DeepSeek has introduced time-based surcharging on its API.
On June 23, 2026, Meta launched its first self-branded smart glasses (from $299) powered by Superintelligence Labs' multimodal Muse Spark model, with a companion app extending Muse Spark to phones.
Meta ships first smart glasses powered by its Superintelligence Labs' Muse Spark model.
In late June 2026, several senior DeepMind researchers left in a single week: Gemini co-lead Noam Shazeer to OpenAI, and Nobel laureate John Jumper plus Jonas Adler, Alexander Pritzel and Arthur Conmy to Anthropic, raising doubts about the lab's ability to stay at the frontier.
As top talent leaves Google DeepMind, some question if the lab can remain at the forefront of AI development.
Markets reacted sharply to the DeepMind departures, with Alphabet pacing for its worst trading day in a year and roughly $225B-$270B wiped from its market cap in the sessions around the exodus.
Alphabet paces for worst day in a year after AI talent exits.
On June 18, 2026, Bell, Cohere, Hypertec and BUZZ HPC announced a three-year, US$220M deal to build a fully domestic Canadian AI compute stack (2,304 Grace Blackwell GPUs) running Cohere's models.
Bell and Canadian AI firm Cohere sign deal to operate models using Bell infrastructure.
In mid-June 2026, DeepSeek closed its first-ever external funding round, ~$7.4 billion at a post-money valuation exceeding $50 billion, backed by Tencent, CATL, and China's state-backed AI fund.
DeepSeek raises $7.4 billion at $50 billion valuation in landmark AI funding round.
OpenRouter's June 2026 open-weight review named Qwen the most important open-model franchise, citing its 1B+ cumulative Hugging Face downloads amid Chinese models reaching ~61% of platform token volume.
Qwen is the most important open-model franchise in the world. It has surpassed 1 billion cumulative downloads on Hugging Face.
June 2026 reporting detailed low morale and multiple senior departures at Meta Superintelligence Labs months after its formation, raising execution questions at Meta's flagship AI unit.
Meta's months-old AI unit is a soul-crushing gulag, say the engineers stuck inside it.
In mid-June 2026, Bloomberg and TechCrunch reported Mistral was in early talks to raise about EUR 3B at a roughly EUR 20B (~$23B) valuation, nearly double its Series C mark. (Reported June 12, just outside the strict window.)
Mistral is rumored to be raising EUR 3B at EUR 20B valuation.
Cohere released North Mini Code, a 30B (3B active) open-weight MoE coding model under Apache 2.0 with a 256K context, its first model aimed at developers and able to run on a single H100. (Announced June 9, just outside the strict window.)
Introducing North Mini Code: Cohere's First Model For Developers.
OpenRouter data showed DeepSeek held nearly 20% of token share and had been the platform's top model since mid-May 2026, with V4-Flash leading aggregate token volume (~10.9T tokens) and ~70% of DeepSeek's agentic usage.
By the start of June, DeepSeek had earned nearly 20% of token share and has been the top model on OpenRouter since mid-May.