OpenAI · Status at source date: Newsletter archive
OpenAI launches GPT-5.5 with sharper agentic capability and 1M token context
OpenAI has released GPT-5.5 (codenamed "Spud"), the company's newest flagship model, available immediately to ChatGPT Plus, Pro, Business, and Enterprise users in ChatGPT and Codex, with API access following the next day on 24 April. The model is positioned as a step change in agentic capability: handling messy, multi-part tasks, planning, using tools, checking its work, and continuing through ambiguity without requiring step-by-step prompting.
What changed
OpenAI president Greg Brockman framed it as "a new class of intelligence" and a step towards "more agentic and intuitive computing".
The improvements are concentrated in agentic coding, computer use, knowledge work, and early scientific research. GPT-5.5 is reported to score 82.7% on Terminal-Bench 2.0 and 73.1% on the internal Expert-SWE benchmark for long-horizon coding tasks. API pricing is $5 per million input tokens and $30 per million output tokens (GPT-5.5 Pro at $30/$180), with OpenAI claiming roughly 40% fewer output tokens per Codex task than GPT-5.4. The release lands six weeks after GPT-5.4: a cadence reflecting active competitive pressure on Anthropic's Claude Opus 4.7.
Why it matters for advertisers
For you, this means GPT-5.5 is now competitive with Claude Opus 4.7 on agentic coding workflows, with the pricing gap narrowing once token efficiency is factored in. Worth re-baselining any model selection decisions made in the past six weeks if your workflows lean towards long-horizon tasks, multi-tool orchestration, or scheduled automation.
Perspective from the original PMC newsletter.Sources & contributor credit
- Newsletter coverage · Paid Media Collective newsletter
Original newsletter text, contributor labels and media for this update.
- Source referenced in newsletterOpenAI
Linked from the original newsletter. The source publication date has not been independently confirmed.
Original creator unverified
The original creator has not yet been verified. Newsletter curation and publication do not establish original authorship.
Attribution evidence and limitations
Full attribution review pending.
A source link alone does not establish original authorship.
Original image and video creators have not yet been verified.
- Published on this site
This update reflects the dated source reporting. Availability may have changed. Further coverage of this same development will be added to this page.
Original newsletter text and archive evidence
OpenAI launches GPT-5.5 with sharper agentic capability and 1M token context
OpenAI has released GPT-5.5 (codenamed "Spud"), the company's newest flagship model, available immediately to ChatGPT Plus, Pro, Business, and Enterprise users in ChatGPT and Codex, with API access following the next day on 24 April. The model is positioned as a step change in agentic capability: handling messy, multi-part tasks, planning, using tools, checking its work, and continuing through ambiguity without requiring step-by-step prompting. OpenAI president Greg Brockman framed it as "a new class of intelligence" and a step towards "more agentic and intuitive computing".
The improvements are concentrated in agentic coding, computer use, knowledge work, and early scientific research. GPT-5.5 is reported to score 82.7% on Terminal-Bench 2.0 and 73.1% on the internal Expert-SWE benchmark for long-horizon coding tasks. API pricing is $5 per million input tokens and $30 per million output tokens (GPT-5.5 Pro at $30/$180), with OpenAI claiming roughly 40% fewer output tokens per Codex task than GPT-5.4. The release lands six weeks after GPT-5.4: a cadence reflecting active competitive pressure on Anthropic's Claude Opus 4.7.
For you, this means GPT-5.5 is now competitive with Claude Opus 4.7 on agentic coding workflows, with the pricing gap narrowing once token efficiency is factored in. Worth re-baselining any model selection decisions made in the past six weeks if your workflows lean towards long-horizon tasks, multi-tool orchestration, or scheduled automation.
Source captured . No explicit first-contributor label was provided for this update.
