Vaibhav (VB) Srivastav(@reach_vb):Big FYI: DO NOT change models unless you really know what you’re doing, the prompts for these models are carefully calibrated for the defaults.

Vaibhav (VB) Srivastav

@reach_vb

founder mode @OpenAI | ex @huggingface | F1 fan | *opinions my own

Joined June 2017

336 Following 62.2K Followers

Vaibhav (VB) Srivastav@reach_vb

2026.10.28 17:22

Big FYI: DO NOT change models unless you really know what you’re doing, the prompts for these models are carefully calibrated for the defaults.

Dan McAteer@daniel_mac8

2026.10.27 15:32

Codex's Dreaming V3 memory system uses GPT-5.6 Luna and Terra by default. Upgrade by copying the following into config.toml: extract_model = "gpt-6-luna" consolidation_model = "gpt-6.1-terra" Both models are more intelligent AND cheaper than their predecessors. It's better than a free lunch. It's like getting paid to eat lunch.