Testing 400K context with GPT-6 Astra in Codex: no extra usage for long input
daniel_mac8 · x · 2026-09-09
Daniel Mac shares a working config for pushing higher context limits in Codex:
toml
model = "gpt-6-astra"
modelcontextwindow = 400000
modelautocompacttokenlimit = 360000
- He reports that Astra doesn't incur higher usage for long-context, making large windows effectively free to try
- Goal is to see how it behaves with Codex's new experimental context management
- Tip credited to @brandongalang
A concrete, copy-paste tweak for anyone running Codex with long-context workflows.
Related event: Developers Test GPT-6 Astra's 400K Context Window in Codex(2 posts)→
More from coding & agent
- Dev Builds Open-Source ARPG Evergrow with Astra single task high, Best Speed-Quality Ratio — Dimillian · 2026-09-09
- Codex + Magnific MCP + PixVerse skill turns one idea into a finished 4K video — aziz4ai · 2026-09-09
- Codex's Underrated 'More Details' Feature Explains Any Text Without Eating Your Quota — jdjohnson · 2026-09-09
- One global function takes LLM-evolved level generation from ~0% to ~100% playability — Amidos2006 · 2026-09-09
- llama.cpp launches llama.app: one-line local LLM install, zero telemetry — ngxson · 2026-09-09
- levelsio vibe-codes $25,000/mo of SaaS away: weather, NSFW detection, streaming, all self-built — NirantK · 2026-09-09