Nikunj Handa
@nikunjhanda
@openaidevs | nikunj.sg@gmail.com ๐ฎ๐ณ๐ด๐ฒ๐ฌ๐ญ๐ธ๐ฌ๐บ๐ธ
1.1K Following 4.5K Followers
cool new responses api features releasing today with astra:
1. async function calling โย model can keep going while your tool runs
2. mid-turn steering โย you can inject messages into the model while it's reasoning to steer it in a different direction (incl. tool outputs for the async function calls)
3. change reasoning effort without breaking cache
Show more
This is GPT-6 Astra.
Anything you can do on a computer, Astra can do for you. Fast.
More capable models need more capable safeguards. Today weโre previewing Private Safety Processing: a new system weโre building to strengthen safety monitoring across interactions while protecting customer privacy and continuing to offer ZDR for our frontier models.
Show more
GPT-5.6 Sol is now 70% off in Devin Desktop and Devin CLI, through October 3.
On FrontierCode 1.1, Sol achieves top-tier results. With this promo, it's one of the most cost-effective frontier models you can run in Devin.
Show more
What is an obvious thing that we should do with Codex, API or our models that we should just do but haven't yet? What is 100% within reach, but we just seem to be missing?
OpenAI's GPT-5.6 Sol is now half off on OpenRouter!
This also applies to batch API, flex, and priority (fast) tiers, giving you as low as $1.25 in and $7.50 out on the flex tier.
Discount applies automatically, just use openai/gpt-5.6-sol.
Show more
GPT 5.6 Sol is 50% off on AI Gateway.
โข Through September 18, 2026
โข Applies to both standard and fast mode
'๐๐๐๐๐๐/๐๐๐-๐ป.๐ผ-๐๐๐'
Show more
Our Black Hat talk on the OpenAI-Hugging Face incident is now live on youtube.
This is a watershed moment for the industry. I encourage all defenders to watch, consider how attack dynamics will imminently change, and plan for accelerating defense.
Show more
OK, GPT-5.6 Luna is a bit of a beast. Given the 80% price drop today I decided to try it in Datasette Agent, and it's furiously quick and generates all the SQL, HTML and JavaScript (for Datasette Apps) I could possibly want
Show more
Updates to GPT-5.6 reasoning context
One of the quieter but more significant GPT-5.6 updates I'm excited to have played a part in is that reasoning context is preserved across turns, giving:
- better KV cache hit rate - reduces latency and cost
- less repeated reasoning on long-running tasks
- stronger follow-up behavior in complex environments when a task depends on prior hidden thinking
Show more
Multi-agent is now in beta in the Responses API!
We highly recommend using websocket mode for it for the best performance, but it also works with most HTTP integrations. All you need to do is set multi_agent.enabled = true.
Try it out in Codex's `ultra` mode. Would love your feedback on this one so that we can make it truly great when we GA this feature:
Show more
Multi-agent lets GPTโ5.6 spawn concurrent subagents within a single request.
They can divide complex work, explore different approaches in parallel, and synthesize their findings into one result.
This feature is initially available in beta via the Responses API.
Show more
Programmatic tool calling (aka code mode) is here for OpenAI models via the Responses API.
We've seen significant gains in token efficiency (lower cost and latency) when using it and because it runs in in-memory V8s, it's highly performant and ZDR-compatible with no additional container costs.
Show more
Programmatic Tool Calling lets GPT-5.6 write and run JavaScript to coordinate complex tool workflows, process results, and decide what to do next.
That means fewer model round trips, fewer tokens, and less orchestration for developers.
Programs run in isolated hosted V8 runtimes, so developers get stateless code-driven orchestration without managing containers.
Show more
we have heard enterprises on their concerns about AI costs, and 5.6 sol is a huge step forward for dollars-per-task, as are terra and luna
GPT-5.6 Sol, along with Terra and Luna, will launch publicly this Thursday.
Weโre expanding preview access globally now.
Show Codex a workflow once. Reuse it as a skill.
Record & Replay lets you show Codex a recurring task, like filing an expense report or submitting a time-off request.
Codex turns that demo into an inspectable, editable skill.
You control when recording starts and stops.
Show more
And it doesnโt stop at answers.
Describe what youโre building and the agent will create a custom guide with a tailored prompt and relevant resources.
Open the guide in Codex or copy it as Markdown for your coding agent.
Show more
We just shipped a search and command bar to the OpenAI API Platform! Use โ๐บ to search through:
โ Pages
โ Settings
โ Quick Actions
โ Resources
โ Developer Docs
Your app can now search the web for images.
Web search in the Responses API now supports image results in addition to text results, so you can build apps that surface products, places, visual references, and source links for inspiration.
Show more
Today we're rolling out a redesigned navigation for the OpenAI API platform, making it easier to find what you're looking for.
Watch me control my computer with just my voice. This is the future of operating systems.
No hands.
GPT-Realtime 2.0 is very, very underrated.
Demo:
Today weโre releasing DeepSWE, a new standard for agentic coding benchmarks.
On public leaderboards, top models often look relatively close in capability. DeepSWE shows where they actually diverge, reflecting the realistic experience of developers in their day-to-day work.
Show more
