Pricing75d ago
OpenAI cuts GPT-5.6 Sol API and credit pricing by over 20% for three months
OpenAI announced that API and credit pricing for GPT-5.6 Sol will drop by more than 20% for the next three months as the company continues to push frontier capability while improving efficiency. The cut is live on the API and rolling out across eligible ChatGPT Work and Codex credit plans. Pro, Plus, and Business subscription usage is unchanged. For teams whose coding agents, eval harnesses, and batch refactors already default to Sol, this is a straightforward unit-cost reduction rather than a new model SKU: the same frontier endpoint becomes cheaper for a bounded promotional window. OpenAI pointed developers to the public pricing docs at developers.openai.com/api/docs/pricing. Engineering orgs should check whether workspace credit pools, Codex usage, and API budgets were hard-coded against the prior rate card, and whether it is now cheaper to keep Sol as the default instead of routing overflow to smaller models.
Pricing54d ago
GitHub Copilot: GPT-5.6 Sol 30% off for Pro+ and Max until Sep 13 UTC
GitHub is offering GPT-5.6 Sol at 30% off inside Copilot for Pro+ and Max subscribers. The promo ends September 13, 2026 at 00:00 UTC, cutting short-window Sol usage cost for heavy coding-agent users.
Research18d ago
Humans Score 100% While GPT-5.6-Sol Hits Only 17%: RoTA-Bench Exposes 3D Interactive Control Gaps, and OpenSAE Pairwise Steering Lifts Fidelity to 0.779
Two top-trending Hugging Face papers dissect the limits of VLM 3D interactive reasoning and LLM mechanistic interpretability: Xiamen University's RoTA-Bench (arXiv:2610.17665, 200 upvotes) evaluates closed-loop 3D-to-2D projection alignment across 674 tasks, finding that humans achieve 100% success while GPT-5.6-Sol reaches only 17%. Meanwhile, UW-Madison's OpenSAE study (arXiv:2610.21261, 104 upvotes) introduces SAE Additivity Residuals (SAR) across 18 model/behavior settings, showing that modeling pairwise feature interactions boosts multi-feature steering rank fidelity from 0.5874 to 0.7790 and cuts ordering error from 0.1444 to 0.0606.
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.