You are likely going to see a lot of very varied results posted online from GPT-5 because it is actually multiple models, some of which are very good and some of which are meh. Since the underlying model selection isn’t transparent, expect confusion.
LLMS
-
OpenAI Dominates Intelligence Frontier with Continuously Trained Router Model
By
–
A lot of developers unimpressed with the GPT-5 numbers are missing the bigger picture: @OpenAI now completely dominates the Intelligence Pareto frontier and for the first time, they have officially confirmed a continuously trained realtime router model and provided some
-
OpenAI GPT-5 Launch: Industry Impact and Recognition
By
–
When @OpenAI asked me to be part of the GPT-5 launch video, it felt like a full-circle moment. GPT-4 changed my life, and I’m grateful I had the chance to give back this time. Proud to have played a small part in GPT history.
-
OpenAI CEO Sama AMA GPT-5 Team Discussion
By
–
AMA with @sama + some members of the GPT-5 team Tomorrow 11am PT.
-
OpenAI’s GPT-5 Series Now Available on Replicate
By
–
OpenAI's GPT-5 series is now on Replicatehttps://t.co/CSH2FmZ4EM pic.twitter.com/v0mFhA5ayS
— Replicate (@replicate) 7 août 2025OpenAI's GPT-5 series is now on Replicate http://replicate.com/openai/gpt-5
-
GPT-5 Pro Mode Chains 100 Instances for Better Outputs
By
–
Introducing `gpt-5 Pro Mode`!
— Matt Shumer (@mattshumer_) 7 août 2025
Basically o3 Pro, but for GPT-5 🙂
Pro Mode chains together up to 100 instances of the new OpenAI GPT-5 model, enabling it to produce a better answer than one instance alone could create!
Open source, link here: pic.twitter.com/mLtM5ZHJ6YIntroducing `gpt-5 Pro Mode`! Basically o3 Pro, but for GPT-5 🙂 Pro Mode chains together up to 100 instances of the new OpenAI GPT-5 model, enabling it to produce a better answer than one instance alone could create! Open source, link here:
-
Early GPT-5 Access and Prompt Testing Experience
By
–
Thanks! I had access to GPT-5 weeks in advance, so I was actually able to test this prompt and use it on a number of tasks 🙂
-
When will OpenAI restore previous models to selector?
By
–
Dicho esto, lanzo la pregunta que he dicho durante el directo. ¿Cuánto tardaremos en pedirle de vuelta a OpenAI los modelos anteriores en el selector?
-
Model Succession: Each AI Must Defeat Its Predecessor
By
–
Before it gets released, every model has to fight and kill its ancestor, or be killed pic.twitter.com/4rI5s1CsbD
— Joscha Bach (@Plinz) 7 août 2025Before it gets released, every model has to fight and kill its ancestor, or be killed