Autonomous coding agent that runs each task in its own cloud sandbox preloaded with your repo — writing features, fixing bugs, running tests, and opening PRs. Reachable from ChatGPT web, a CLI, desktop apps, and IDEs (VS Code, JetBrains, Xcode).
Zirui Guo, Lianghao Xia +3University of Hong Kong, Beijing University of Posts and Telecommunications
A graph-based RAG framework pairing a knowledge graph with vector retrieval and a dual-level (low/high) query mode. New documents merge into the graph via set operations instead of triggering a rebuild, cutting the cost of keeping the index current.
AI API2020
Lets developers build AI features with hosted frontier models for text, code, vision, audio, images, and agents. The platform pairs model APIs with tools, SDKs, safety controls, and enterprise options.
AI Client2025
Manages provider configs for seven coding CLIs (Claude Code, Codex, Gemini CLI, OpenCode and more) from one desktop app, so switching API endpoints no longer means hand-editing JSON, TOML, or .env files. Adds tray quick-switch and cloud sync.
Introduced the two-stage recipe behind the GPT lineage: unsupervised generative pre-training on unlabeled text, then supervised fine-tuning per task. A single 12-layer Transformer decoder beat bespoke architectures on 9 of 12 NLP benchmarks.
A 1.5B-parameter model trained only to predict the next token on diverse web text does translation, summarization, and QA zero-shot, with no fine-tuning. It recast NLP tasks as conditional language modeling and sparked the staged-release misuse debate.
Demonstrated that language model loss falls as a smooth power law in model size, data, and compute across more than seven orders of magnitude — turning "make it bigger" from a hunch into a budget you can plan, and justifying the GPT-3 scale-up.
At 175 billion parameters, this autoregressive model becomes a strong few-shot learner: it handles translation, QA, and reasoning from a few prompt examples with no gradient updates, establishing in-context learning as an alternative to fine-tuning.
AI API2020
Typed Python client for the OpenAI REST API that offers synchronous and asynchronous clients, typed request/response models, streaming and Realtime support, webhook verification, and integrations for Azure and Amazon Bedrock—built for production integrations and automation.
Showed that fine-tuning a GPT model on public GitHub code yields a capable program synthesizer, and introduced HumanEval — the docstring-to-function benchmark that still anchors code-generation evaluation. A production variant powers GitHub Copilot.
Made reinforcement learning from human feedback (RLHF) the standard alignment recipe: collect demonstrations and preference rankings, train a reward model, then optimize with PPO. A 1.3B aligned model was preferred over the 175B GPT-3 by human raters.