daily notes

2026-10-01

Themes

GPT-6.1 Sol shifts the disclosed cost/capability tradeoff

OpenAI presents GPT-6.1 Sol as near-Astra capability at one-fifth of standard input/output token prices. Artificial Analysis reports $0.72 versus $3.26 per Intelligence Index task at max effort, but its benchmark is a comparator rather than a universal usage result.

ArtificialAnlysoriginal post ↗
GPT-6.1 Sol pushes the cost efficiency frontier for OpenAI models. OpenAI has also fixed an issue reading multimodal input which improves Intelligence Index results for GPT-6 Luna by 1 point. At max effort, GPT-6.1 Sol costs less than a quarter of GPT-6 Astra per Intelligence Index task ($0.72 vs $3.26). It also costs 31% less per task than GPT-6 Sol ($1.04) and 64% less than GPT-5.6 Sol ($1.99). All effort levels of GPT-6.1 Sol push out the cost efficiency Pareto frontier: for a given level of intelligence, there is no cheaper model. OpenAI also fixed a bug in image encoding for GPT-6 Luna and Sol. GPT-6 Luna (max) has gained a point in evals that benefit from visual understanding, while GPT-6 Sol shows negligible improvement.

OpenAI: Introducing GPT-6.1 Sol

Must read

Addendum to GPT-6 Astra System Card: GPT-6.1 SolOpenAI

The release page is useful for cost and capability claims; this addendum is the deeper read. It documents the safety-evaluation frame, inherited safeguard stack, and limitations, including contamination risk and areas where the model does not reach the higher threshold. Read it to separate launch claims from the evidence and boundaries that govern deployment.

OpenAI DevDay 2026: Dots, 6.1 Sol, Ultrafast, Decisions API, Agents API, Spaces, Marketplace, and 1.2 Billion ChatGPT WAUAINews

A long-form synthesis that connects the DevDay releases instead of treating them as isolated launch bullets. Its value is the implementation and product context around Dots, decisions, APIs, Spaces, and model economics, with primary links available for readers who need to inspect the underlying claims.

Sentiment

A disclosed cost-capability change has external benchmark support, with workload limits still material +0.20

37 source items · 84% editorial confidence

Must read

Core Principles for Effective Deposit Insurance Systems – Executive SummaryInternational Association of Deposit Insurers

This revised international benchmark makes deposit insurance a framework problem rather than a coverage-limit debate: legal powers, funding, supervision, resolution, and cross-border coordination all have to work together. The executive summary is short enough to read in one sitting but substantial enough to sharpen how a reader evaluates bank-resolution resilience.

Signals

August trimmed-mean PCE eased to 2.2%Dallas Fed

DallasFedoriginal post ↗
The Trimmed Mean PCE inflation rate over the 12 months ending in August was 2.2 percent, down slightly from July, according to the Dallas Fed. The one-month annualized rate of Trimmed Mean PCE inflation was 1.9 percent. See the full breakdown here: https://t.co/U9EgAVcWfq

Dallas Fed's 12-month trimmed-mean PCE reading for August was 2.2%, down from 2.23% in the prior June coverage; its one-month annualized reading was 1.9%. This is an alternative underlying-inflation measure, not a replacement for the BEA release's 3.4% headline and 3.0% core PCE readings.

updates 2026-07-30 · prior evidence ↗

Sentiment

Underlying inflation eased at the margin while the broader PCE measure remains elevated +0.02

143 source items · 81% editorial confidence