Sign in Subscribe

Bob M

Cognition's $40B Devin Valuation Talks. Here's The Math.

Cognition's $40B Devin Valuation Talks. Here's The Math.

Cognition's Devin just landed in $40 billion valuation talks. The San Francisco startup behind the AI coding agent is reportedly in early talks to raise over $1 billion at a valuation of at least $40 billion, Bloomberg first reported and TechCrunch surfaced on August 12, 2026. That would

DeepSeek V4 Pro 0813 Ships to GA: Pricing, Benchmarks, and the Quiet Rollout

DeepSeek V4 Pro 0813 Ships to GA: Pricing, Benchmarks, and the Quiet Rollout

Sometime around 11 p.m. Beijing time on August 13, 2026, the `deepseek-v4-pro` endpoint quietly became DeepSeek V4 Pro 0813 and nobody sent a memo. The model had been living in preview for nearly four months. Then it graduated to general availability. Chat interface picked it up, API picked it

Grok Bot Shipped. Your New AI Teammate Logs In As You

Grok Bot Shipped. Your New AI Teammate Logs In As You

Tuesday, August 11, 2026. SpaceXAI shipped Grok Bot, and it's the first real answer to what the SpaceXAI-Cursor merger was actually for. An always-on AI teammate that gets its own cloud computer, logs into your apps. And finishes jobs while you sleep. Not text. Work. Grok Bot Features

Pathway's 150M BDH-CQ Broke the ARC-AGI-1 Cost Frontier

Pathway's 150M BDH-CQ Broke the ARC-AGI-1 Cost Frontier

Pathway's BDH-CQ, a 150M-parameter post-transformer reasoning model, scored 29.5% pass@2 on the public ARC-AGI-1 evaluation set at a computed inference cost of $0.0007 per task. That is roughly 11 times cheaper per task than GPT 5.6 Luna (Low), even after OpenAI cut Luna'

Prime Agent by Prime Intellect: A Self-Modifying AI Coding Scaffold

Prime Agent by Prime Intellect: A Self-Modifying AI Coding Scaffold

Started wrong. Spent a paragraph on the benchmark number before realizing I was staring at the wrong end of the thing. 95.5% RHAE Best@1 on ARC-AGI-3. Opus 5 inside the agent. Prime Intellect's reported figure, relayed through TestingCatalog. A tenth of a point above the 95.

Qwen3.8-Max Open Weights Drop Next Week. First Time Ever.

Qwen3.8-Max Open Weights Drop Next Week. First Time Ever.

Qwen3.8-Max landed on August 3, 2026. And the part most people skimmed past is the part that changes everything: Alibaba is opening the weights. A frontier-class Chinese model, downloadable. That hasn't happened before. Qwen3.8-Max Specs and Pricing Here's what you're actually getting

Muse Glimmer: Meta Open-Sources A 30B Local Agent Model

Muse Glimmer: Meta Open-Sources A 30B Local Agent Model

Muse Glimmer is Meta Superintelligence Labs' new 30-billion-parameter model, released under Apache 2.0 and built to run on a Mac or PC with a single consumer GPU. Meta distilled it from Muse Spark (Bloomberg pegs the source as Muse Spark 1.2) and shipped it for always-on local