Here's a quick recap of the AI news I came across this week: – OpenAI launched #ChatGPTAgent. I put it to the test in my new video out tomorrow
– Moonshot released open-source Kimi K2
– Anthropic launched Claude Tool Directory for app integrations
– Anthropic announced Claude
LLMS
-
OpenAI ChatGPT Agent, Kimi K2, and Claude Tool Directory launches
By
–
-
OpenAI Tests Anonymous Chatbot 0717 Web Development Model
By
–
OpenAI are testing a new model on the Web Dev Arena @lmarena_ai under the name 'Anonymous Chatbot 0717'. I can't believe I'm gonna say this, but it is genuinely at a completely different level of front end coding – far better than Sonnet, o3, Gemini 2.5 Pro, or Grok 4.
— Peter Gostev (@petergostev) 18 juillet 2025
To test… pic.twitter.com/wQKMgPRFGFOpenAI are testing a new model on the Web Dev Arena @lmarena_ai under the name 'Anonymous Chatbot 0717'. I can't believe I'm gonna say this, but it is genuinely at a completely different level of front end coding – far better than Sonnet, o3, Gemini 2.5 Pro, or Grok 4. To test
-
Wish for Technical Report on GPT-4O Attention Deficiencies
By
–
man i wish i had a technical report that specifically documented how badly gpt 4o is deficient in attention
-
AI Dot Engineer Conference Daily Track Releases and MCP Coverage
By
–
we're releasing one track a day from the @aidotengineer conf now*. yesterday's RecSys track was a big hit – but by far the hottest track was our coverage of the state of MCP, hosted by @Calclavia personal fave slide is this where i realized @AnthropicAI dogfoods MCP -way-
-
Llamafile enables multi-platform LLM deployment with single binary
By
–
llamafile is pretty great for that, since one binary will work on multiple different OS and platforms
-
Persuasion techniques tested on GPT-4o model
By
–
I didn't realize that Bob had an X account: @RobertCialdini And we did test GPT-4o as well and found that persuasion worked for that model as well, when there weren't floor or ceiling effects.
-
Hardware independent LLM inference engine from ZML
By
–
Hardware independent LLM inference engine from ZML.
-
State-of-Art AI Models Made Accessible to Developers
By
–
Excited to bring SoTA models closer to developers with you! Thanks for the support and sprint!
-
Groq Launches Kimi K2 Model for Developer Building
By
–
1 week ago: Kimi K2 launches
72 hours later: We YOLO launch it on Groq
Now: Thousands of devs are building with it Kimi K2, now on Groq: 1T parameters Full context Built for agents Unmatched price-performance Build Fast. Link in Comments. -
Using Grok for prompt-based analysis
By
–
copy paste the full prompt into Grok 4, answer the questions