OpenAI Gives 100,000 Researchers Free GPT-5.6 Access

OpenAI announced on July 29 that it is launching ChatGPT for Academic Researchers, a program that will extend free access to its most capable models — including GPT‑5.6⁠ Sol Pro — to 100,000 scientists, mathematicians and engineers. The rollout begins with 10,000 participants this summer, with expansion to the full 100,000 planned through 2027.

Early access is already live at the Institute for Advanced Study and École normale supérieure. Applications are open now to researchers at other qualifying institutions, which must be recognized, degree-granting universities with a high level of research activity. Applicants must verify their institutional affiliation and describe their active research before being approved.

What the Program Actually Includes

Approved researchers receive access to the full GPT‑5.6⁠ family across ChatGPT, ChatGPT Work and Codex. The three tiers are designed for different workloads: GPT-5.6 Luna handles lighter tasks quickly, GPT-5.6 Terra balances capability with efficiency for everyday research, and GPT-5.6 Sol is positioned for the hardest scientific and mathematical problems. On FrontierMath Tier 4, a benchmark for research-level mathematical reasoning, GPT-5.6 Sol scores 83%, compared with 72.5% for its predecessor GPT-5.5. On GeneBench Pro, which tests complex biological data analysis, GPT-5.6 Sol Pro solves 31.5% of tasks.

Beyond raw model access, participants get more than 75 life science skills spanning genetics, genomics, protein modeling and drug discovery, plus connectors to scientific literature databases, public genomic and clinical repositories, satellite imagery, computational notebooks, and reference managers. Codex supports code writing, debugging and reproducible workflow construction. ChatGPT Work is aimed at longer-horizon tasks like grant preparation, literature reviews and manuscript drafting. Workspaces include business-grade privacy protections, and data is not used to train OpenAI’s models by default.

Each approved researcher can invite up to four collaborators from their institution — all of whom must verify their affiliation independently.

The Bigger Financial Picture

The academic researcher program sits inside a broader financial commitment OpenAI describes as more than $250 million through 2027 for external scientific research. That umbrella also includes NextGenAI, a $50 million initiative supporting research institutions, and a collaboration with the U.S. Department of Energy’s Genesis Mission⁠, which aims to bring frontier AI to national laboratories and universities.

OpenAI says roughly 1.3 million people per week already use ChatGPT for advanced science and mathematics, producing around 8.4 million messages weekly. The company notes that researchers in the top 20% of AI usage within their field were almost twice as likely as peers to assign tasks estimated to require four or more hours of human work — nearly 7% of their requests, versus 3.5% among other researchers in the same fields.

Where This Fits in a Fragmenting Competitive Landscape

The announcement lands at a moment when free frontier AI access for researchers and students has become genuinely scarce. Google’s free year of Gemini Advanced for students ended globally, with the final U.S. redemption window closing in April 2026; new students now get only the free Gemini tier or a one-month Pro trial. Anthropic has no publicly redeemable student discount for Claude. OpenAI itself does not offer a standing student discount for ChatGPT. Microsoft Copilot provides a three-month free trial followed by a 50% ongoing discount for students, but that is a consumer productivity product rather than a research-grade AI environment.

On the model capability side, the picture is more nuanced. GPT-5.6 Sol leads on agentic benchmarks like Terminal-Bench 2.1 and BrowseComp, but Claude Fable 5 still outperforms it on pure code generation via SWE-Bench Pro, 80% to Sol’s 64.6%. OpenAI’s program is widely read in the industry as a strategic move to establish a foothold in higher education: departments that adopt the tool collectively could create pipeline pressure toward paid ChatGPT subscriptions and API usage down the road.

What This Means If You’re a Graduate Student

The program is formally aimed at faculty and postdoctoral researchers, not undergraduates — but the collaborator-invite model creates a concrete pathway for doctoral and master’s students. If your adviser or principal investigator qualifies and gets approved, they can add you as one of their four institution-linked collaborators. That would give you access to GPT-5.6 Sol Pro, a model that carries a cost of $5 per million input tokens on OpenAI’s open API, at no personal expense.

For students in STEM fields, that access could meaningfully change what is feasible on thesis work, grant preparation and literature synthesis without running up a personal software bill. More broadly, the usage data OpenAI published reinforces a point worth taking seriously: the researchers who are integrating AI most deeply are tackling more ambitious problems, not just automating routine tasks. For students entering STEM careers, knowing how to deploy these tools for complex, multi-hour workflows — rather than quick queries — is increasingly what separates candidates in a competitive job market.

Applications for ChatGPT for Academic Researchers are open now. Read OpenAI’s full announcement and review eligibility criteria here.

Source: OpenAI