first: GPT OSS is launched with the harmony format https://
openai.com/open-models/ https://
github.com/huggingface/tr
ansformers/releases/tag/v4.55.0
…
@swyx
-
GPT OSS Launch with Harmony Format and Model Updates
By
–
-
Why I Chose Tiny Teams for AI Engineer and Fanahova
By
–
then: https://
x.com/swyx/status/19
26410472590385413
… i landed on Tiny Teams for @aiDotEngineer and [upcoming @fanahova project] because: 1) @awilkinson (who sadly had to drop out) 2) our @sidbendre chat on @latentspacepod 3) wanted to pick a more constructive direction for the industry than Vibe -
AI Engineer Thesis: GPT Wrappers Outvaluing Research Labs
By
–
the ai engineer thesis is finally happening!!!
gpt wrappers more highly valued than labs!!! harvey: 50x ARR (
https://
x.com/brian_a_burns/
status/1952400121926439336
… vs 5b valuation) who else -
OpenAI’s Universal Verifier Extends AI Capabilities to Subjective Domains
By
–
OpenAI has developed a "universal verifier" that could help translate its gains in domains like math and coding to other, more subjective domains like business decision-making or creative writing. We have the details here: https://
theinformation.com/articles/unive
rsal-verifiers-openais-secret-weapon
… -
Claude Code Compaction Methods and Optimization Challenges
By
–
good list of claude code tips. it's surprising how often i hear about compacting being the weak point of CC – has anyone written a hook/hotfix or explored different methods of compaction here?
-
AI Valuation Multiples: OpenAI vs Ant Financial Analysis
By
–
current multiples are interesting. OAI: 15x EOY25 ARR
Ant: 19x EOY25 ARR details
OAI 300b valuation, 13b ARR now, 20b EOY
Ant 170b valuation, 5b ARR now, 9b EOY these don't seem… crazy? -
DeepThink vs O3 Pro: Dramatic Performance Gap Analysis
By
–
not enough people are talking about the delta between the parallel thinking uplifts of oai vs gdm AIME
o3 pro: +3% (from 90->93 on 2024)
deep think: +11.2% (from 88->99.2 on 2025) Knowledge
o3 pro: +3% (on GPQA)
deep think: +13.2% (on HLE) Coding
o3 pro: +9.1% (on Codeforces -
Collecting Examples of Defying Algorithm Logic
By
–
i have been collecting "fuck you algorithm" examples
-
ChatGPT for Meme Making: First Attempt and Challenges
By
–
first time trying using pure chatgpt for meme making, surprised it kinda sorta worked, left in the artefacts for interest. its really hard to get an llm to be funny.
-
Perplexity Finance reliability compared to 3B language model
By
–
perplexity finance’s numbers have the reliability of a 3b llm btw still unserious (for now)