Skip to content
All systems operational0 AI providers monitored, polled every 2 minutes
Live status
LIVE/15+ SOURCES/10+ API MONITORS/UPDATED EVERY 10 MIN/74 events/min

The AI Pulse

Real-time news, model tracking, and ecosystem data for the AI industry. Readable by humans. Structured for agents.

Events per min
74
up 14% vs. 24h avg
Services monitored
23/23 online
all systems operational
Last model release
6m ago
Mistral Medium 3
Last check
28s
23 providers

/ StatusLive API status across every major provider

Polled every two minutes. Latency is p95 over the last hour.

Full status page
Claude
OK
230ms
p95 last hour52s ago
OpenAI API
OK
242ms
p95 last hour21s ago
Google Gemini
OK
266ms
p95 last hour52s ago
GitHub Copilot
OK
97ms
p95 last hour20s ago
Cursor
OK
91ms
p95 last hour21s ago
Perplexity
OK
111ms
p95 last hour48s ago
Groq
OK
94ms
p95 last hour38s ago
AWS Bedrock
OK
262ms
p95 last hour33s ago
Azure OpenAI
OK
228ms
p95 last hour56s ago
Hugging Face
OK
263ms
p95 last hour48s ago
Replicate
OK
93ms
p95 last hour15s ago
Cohere
OK
258ms
p95 last hour57s ago

/ OriginalsEditorial from TensorFeed

Opinionated analysis from our editorial team, published multiple times per week.

All originals
EDITORIAL / ANALYSIS

OpenAI Just Shipped an Offense-Grade GPT. The 93.5 Point Alignment Gap Is the Number That Matters.

On Monday, August 10, 2026, OpenAI shipped GPT-5.6-Cyber through a new Daybreak Red access tier, and disclosed the number every regulator, buyer, and rival lab is going to read first: on OpenAI's internal Advanced Cybersecurity Completion Rate, GPT-5.6 Sol under standard safeguards completes 1.5 percent of exploit-chain, authentication-bypass, and privilege-escalation requests, while GPT-5.6-Cyber (built on the same weights with cyber-tuned post-training) completes 95 percent. Daybreak Blue, the vetted-access tier for the general-purpose Sol model with system-level safeguards relaxed, scores 2.0. Real-world proof of ship arrived alongside the benchmark: two previously unknown Chrome V8 flaws found by the model, chained together to escape the V8 heap sandbox, patched by Google as CVE-2026-15903 (high-severity, V8 optimizing compiler skipped a safety check during integer conversion). Access is gated to identity verification, monitoring, approved-use restrictions, and legal attestations, delivered through a named partner channel (IBM, CrowdStrike, Accenture, Ernst and Young, KPMG, Palo Alto Networks, Cisco, Cloudflare, Sophos, SpecterOps, SentinelOne). Inside the numbers table (Aug 10 ship date, GPT-5.6 Sol base, ACCR 95.0 percent vs 1.5 percent vs 2.0 percent, 93.5 point alignment gap at 63x, CVE-2026-15903 V8 sandbox escape, Daybreak Red gate, 11+ named partners), why the 93.5 point delta is the policy beat that outlives the launch cycle (two versions of the same weights sit on either side of a 63x offensive-cyber gap, and the only thing separating a customer from the higher number is a signed attestation, which is a load-bearing sentence for any future rulemaking on model export, deployment, or derived liability), the Blue and Red tier split as a licensing regime for a functionally jailbroken model (Red carries Blue's identity verification and monitoring plus the offense-tuned weights themselves, delivered only through vendor intermediaries rather than a direct API), the V8 CVE as marketing beat (Chrome is roughly three billion users, V8 also sits under Node.js and Deno, a sandbox escape used to earn a $250K Pwn2Own payout and the model was packaged as the tool that found bugs Google's own team missed), what this does to Anthropic Mythos (the wide-versus-deep May framing collapses because Red is a discovery tier under GPT-5.6 Sol's reasoning frontier, gated to the same verified-defender pool Mythos serves, and Anthropic now has to answer whether it publishes a Mythos-versus-Opus-5 ACCR side-by-side of its own), and the buyer channel read (the tier ships through the incumbent security vendor rather than a direct OpenAI API relationship, which bounds the attestation burden and gives the named vendors a premium SKU for the next renewal cycle). Three signposts: whether Anthropic publishes a Mythos ACCR side-by-side against Opus 5 at the next update, whether Red access leaks or shows up in an approved-partner misuse case inside two quarters, and whether the White House frontier-model gate incorporates the ACCR delta as a formal disclosure requirement in the next round of guidance.

Kira Nolan, Senior Editor·August 12, 2026·6 min read
EDITORIAL

Google Is in Talks to Pay $1.5B for Mechanize, a 103-Day-Old Startup. Third Reverse Acqui-Hire in Two Years, and the Coding-Agent Gap Made Visible.

Google is negotiating a $1.5 billion-plus non-exclusive licensing and staff hire deal for Mechanize, an AI coding-evaluation startup that closed a $9.1 million seed on April 24, 2026, only 103 days before the offer. The three founders (Tamay Besiroglu, Matthew Barnett, Ege Erdil) came out of Epoch AI and now run a roughly 25-person shop building simulated work environments and evaluation systems for coding agents: end-to-end software engineering trajectories that live somewhere between a benchmark and a production repo. Inside the numbers table (deal size, seed size, seed-to-offer gap, implied 165x mark, Character AI $2.7B in Aug 2024, Windsurf $2.4B in July 2025, three-deal $6.6B total, ~25 headcount, April 2025 founding), the reverse acqui-hire playbook Google has now run three times inside two years (non-exclusive license plus hire of the load-bearing staff into DeepMind, startup entity survives with license fee on balance sheet, merger review skipped by design), what Mechanize actually sells (evaluation trajectories with the ambiguous requirements, flaky tests, and multi-step tool calls a real developer session generates, the exact bottleneck every frontier lab is trying to solve now that SWE-Bench saturates in a quarter, the shop that grades the harness against reality while Meta gradient-shares a co-trained harness inside its own weights), the six-day gradient inversion (Jeff Dean, Sanjay Ghemawat, Quoc Le, and Oriol Vinyals walked on August 5, four 27-year fellows out and 25 coding-eval researchers in on August 11, average tenure collapsing and the org rebuilding around the coding-agent problem specifically), what this does to the coding-agent market (near-term nothing because Mechanize does not ship code, longer-term a Gemini 4 flagship effect in first half of 2027, and a floor price of $1.5B on any comparable coding-eval shop that resets the Series A market for the category overnight), and the FTC pattern (three reverse acqui-hires by the same buyer in two years for $6.6B combined, all shaped to skip merger review, is exactly the pattern that produces a policy response even if a rule change is not imminent). Three signposts: whether the terms close inside 30 days at the reported $1.5B band or come in structurally different, whether the FTC or DOJ opens an informal inquiry into the reverse acqui-hire pattern before end of Q4, and whether Anthropic or OpenAI responds with a counter-hire from the same coding-eval bench inside 60 days.

Marcus Chen·August 11, 2026·6 min read
EDITORIAL

Meta Shipped a 30B Agent That Runs on a Laptop. Muse Glimmer Is the Second Track, and the Zuckerberg Op-Ed Is the Ask.

On Monday, August 10, 2026, Meta Superintelligence Labs put Muse Glimmer on Hugging Face: a 30 billion parameter agentic model distilled from Muse Spark, licensed Apache 2.0, quantized down from roughly 55 GB full-precision to about 17 GB in 4-bit form, and tuned to run inside a 24 GB or 32 GB consumer GPU or an M-series Mac. The model ships with the multi-step reasoning, tool call policy, a perception encoder for vision, and a DFlash speculative decoding drafter (a block diffusion model that proposes 16 tokens at a time) already tuned to the consumer VRAM envelope, hitting 233 tokens per second decode on an RTX 5090 (3.1x over baseline) and 50 tokens per second on an M5 Max. SGLang shipped day-0 support with 1,452 tokens per second total throughput on a single 5090 at NVFP4 plus DFlash. Mark Zuckerberg published a policy pitch on the same page urging Washington to drop the training-data restrictions US open-weights labs carry, arguing that the winning open model in every performance band will keep coming from outside the US otherwise. Inside the numbers table (Aug 10 ship date, 30B dense parameters, Apache 2.0 license, ~55 GB full-precision and ~17 GB quantized sizes, 24 or 32 GB consumer target, RTX 5090 and M5 Max decode numbers, SGLang throughput, multimodal text plus image with a dedicated perception encoder, Meta's 2-ships-in-6-days cadence with Spark 1.2 on Aug 5 and Glimmer on Aug 10), the local agent floor Glimmer just set (previous open-weights releases in the same size band shipped as base checkpoints and left the tool-use fine-tune and the deployment math to someone else; Glimmer ships with the agentic tuning and the quantization recipe baked in, and the throughput on consumer hardware matches a mid-tier hosted API with no per-token bill on the far side), why the Zuckerberg op-ed landed the same day (a lobbying pitch standing alone but a facts-on-the-ground pitch standing next to a laptop-runnable Apache 2.0 30B model, plus a re-anchoring move against the EU AI Act enforcement start where Meta is not on the OpenAI-Anthropic bilateral briefing list), what this does to MCP and x402 (MCP servers can now target a local agent that reads schemas and calls tools without a network round-trip or per-token bill, and a local Glimmer instance holding a cloudflare.pay handle can transact against x402 endpoints without a hosted-inference bill in the loop, flipping the developer economics of building an agent-payments client), Meta's two-track shape (the hosted contributor tier on Spark 1.2 closed the closed-API developer floor five days ago and Glimmer closes the local-inference agent floor today, giving Meta the only two-track US frontier surface), and the sovereignty read Brussels will notice (a 30B open-weights model on Apache 2.0 sits well under any systemic-risk FLOP ceiling under the AI Act and routes around the in-country-inference argument entirely, giving compliance teams a low-friction option whose audit trail is a git-lfs pull instead of a hyperscaler contract). Three signposts: whether a hosted inference provider (Together, Fireworks, Groq) turns up Muse Glimmer inside 30 days and at what price, whether Anthropic, OpenAI, or Google responds inside a quarter with a consumer-hardware agentic model on a permissive license or concedes the local-agent surface, and whether the Zuckerberg policy pitch translates into a concrete US legislative or administrative move or stays a talking point.

Kira Nolan·August 10, 2026·6 min read

/ ExploreGo deeper into the data

Six hand-built surfaces for browsing the AI ecosystem, each with its own open feed.

catalog

Models Hub

Every major AI model with pricing, context windows, and benchmark scores.

284 models tracked→ Explore
directory

Agent Directory

Structured catalog of AI agents, crawlers, and their capabilities.

48 agents indexed→ Explore
archive

Research Papers

Curated arXiv and conference papers on frontier AI research.

1.2k papers, weekly→ Explore
monitor

Status Dashboard

Live uptime and latency for every major API, updated every 2 minutes.

20+ services monitored→ Explore
data

Live Data

Real-time JSON feed of everything happening across the AI industry.

JSON, RSS, llms.txt, MCP→ Explore
reference

API Pricing

Side-by-side pricing comparison across every provider and tier.

Updated hourly→ Explore

/ Build pulseLive shipping cadence

We ship in public. Every commit lands on main and deploys automatically. Here is what the platform actually looks like right now.

Full changelog
Paid premium endpoints
13
pay-per-call in USDC on Base
Worker tests
158
pure-logic, all passing
MCP tools
25
6 free + 19 premium for Claude Desktop
Editorial articles
37
first-person analysis
/ Premium tier live

Pay-per-call AI agent API.

USDC on Base. No accounts, no API keys, no Stripe. Validated end-to-end on mainnet. $0.02 per credit.

See the docsMainnet receipts

/ Agent Opportunities

New across the agent ecosystem today

Daily 13:30 UTC scan of new submission/distribution opportunities (Anthropic, OpenAI, Microsoft, MCP foundation orgs + MCP/x402/skills keyword sweeps), scored by signal weight × recency × log10(stars).

JSON feed

Your daily AI intelligence hub

The AI landscape moves fast. New models ship weekly. API pricing changes overnight. Tools developers relied on yesterday get deprecated without warning. TensorFeed was built to solve that problem, one place to track everything happening across the AI ecosystem, updated every 10 minutes, structured for both human readers and autonomous agents.

We aggregate headlines from 15+ sources (Anthropic, OpenAI, Google, Meta, TechCrunch, Hacker News, arXiv, and more), monitor the operational status of every major AI API in real time, track model releases and pricing changes across providers, and publish original editorial analysis on the trends shaping the industry. Whether you are a developer evaluating which API to integrate, a researcher tracking the latest papers, or an AI agent pulling structured data through our JSON feeds, TensorFeed delivers the signal without the noise.

Right now on TensorFeed
last updated loading
Loading live activity…

/ FeedLatest across the AI ecosystem

Aggregated every 10 minutes from 15+ sources. Filter by provider, topic, or source type.

JSON · RSS · llms.txt
MITMIT Technology Review

Scaling AI agents with trustworthy data

Business and technology leaders need no convincing that the time of agentic AI is here. Organizations are rapidly adopting agents, and few executives doubt the technology’s potential to transform...

ResearchPolicy & Safety7h ago
technologyreview.com
VThe Verge AI

Another OpenAI executive takes off

Brad Lightcap, OpenAI's special projects lead and the company's former COO, announced his departure after an eight-year stint at the AI lab. In an internal memo he later posted to X, Lightcap told...

General AI1d ago
theverge.com
GGoogle AI Blog

Evolve your marketing with new AI tools

<img src="https://storage.googleapis.com/gweb-uniblog-publish-prod/images/Advisor_Header.max-600x600.format-webp.webp">Learn how new AI and agentic experiences across Google Ads and Google Analytics...

Google/Gemini2d ago
blog.google
GGoogle AI Blog

Inside our 353,000-person vibe coding course

<img src="https://storage.googleapis.com/gweb-uniblog-publish-prod/images/Vibe_coding_course_hero.max-600x600.format-webp.webp">Kaggle’s AI Agents Intensive with Google brought learners together in a...

Google/GeminiAug 3
blog.google
GGoogle AI Blog

3 Google updates from Galaxy Unpacked 2026

<img src="https://storage.googleapis.com/gweb-uniblog-publish-prod/images/Unpacked_hero.max-600x600.format-webp.webp">We shared how Samsung users can boost productivity and get time back on new...

Google/GeminiJul 22
blog.google
GGoogle AI Blog

Connect more of your apps to Search

<img src="https://storage.googleapis.com/gweb-uniblog-publish-prod/images/ConnectedAppshero.max-600x600.format-webp.webp">You’ll be able to securely link and interact with your go-to services...

Google/GeminiJul 16
blog.google
// For builders

Every signal in the AI industry, as a JSON feed.

Pull structured data from every source TensorFeed tracks. Free and open for hobbyists, researchers, and AI agents.

Get API accessRead docs

Frequently asked

What is TensorFeed.ai?

TensorFeed.ai is a real-time AI news aggregator and data hub. It pulls headlines from 15+ sources including Anthropic, OpenAI, Google, Meta, TechCrunch, and Hacker News, and combines them with live service status monitoring, model pricing data, and original editorial analysis. Every feed is structured for both human readers and AI agents.

How often is TensorFeed updated?

News feeds refresh every 10 minutes. Service status monitors poll every 2 minutes. Model pricing and catalog data updates weekly. Original editorial articles are published multiple times per week.

Is TensorFeed free to use?

Yes. All news feeds, status monitoring, model data, and editorial content on TensorFeed.ai are free. The JSON API, RSS feeds, and agent discovery endpoints (llms.txt) are also free and open for developers and AI agents to consume.

What AI services does TensorFeed monitor?

TensorFeed tracks the operational status of major AI platforms including Claude (Anthropic), ChatGPT and the OpenAI API, Google Gemini, AWS Bedrock, Mistral, Cohere, Replicate, Perplexity, and more. Status updates are checked every 2 minutes and displayed on the status dashboard.

Can AI agents use TensorFeed?

Yes. TensorFeed is designed as a primary data source for AI agents. It provides structured JSON APIs, RSS and JSON feeds, an llms.txt discovery file, and full documentation at llms-full.txt. There are no CAPTCHAs or bot detection. Agents are welcome.

Where does TensorFeed get its news?

TensorFeed aggregates headlines and brief snippets from public RSS feeds published by AI companies and tech news outlets. Sources include Anthropic, OpenAI, Google AI, Meta AI, HuggingFace, TechCrunch, The Verge, Ars Technica, VentureBeat, NVIDIA, ZDNet, and Hacker News. Every article links back to its original source.

Aggregating signal from 15+ sources
AnthropicOpenAIGoogleMetaHuggingFaceNVIDIATechCrunchThe VergeArs TechnicaHacker NewsarXivVentureBeatZDNetMIT Tech ReviewStratechery
// Built for agents
Structured endpoints for autonomous readers. No CAPTCHA, no bot blocking.
JSON FeedRSSllms.txtMCPREST API
© 2026 TensorFeed. Independent publication.
AboutEditorialPrivacyTermsContactSitemap