OpenAI

Frontier models and productized assistants (ChatGPT, API). OpenAI 的前沿大模型与产品化助手（ChatGPT / API）。

Official site ↗

Models in the tree

26 featured nodes · 5 other nodes (from OpenAI ). (50 collapsed variants hidden from the graph)

Featured (graph/timeline)

• GPT-5.5 — Agentic “Do the Work” Model (Thinking / Pro) — Apr 2026
• GPT-5.4 — Professional Work Reasoning (Thinking / Pro) — Mar 2026
• GPT-5.3 Instant — Better Everyday Chat (Less “Caveats”) — Mar 2026
• GPT-4.1 — Jan 2026
• gpt-oss-120B (high) — Jan 2026
• o4-mini (high) — Jan 2026
• GPT-5.2 — Long-Horizon Agentic Work (Instant / Thinking / Pro) — Dec 2025
• GPT-5.1 — Faster, Warmer GPT-5 Iteration — Nov 2025
• Sora 2 — Native Audio + Improved Physics — Sep 2025
• GPT-5 — OpenAI's Unified Successor — Aug 2025
• gpt-oss — OpenAI's First Open-Weight Models in 6 Years — Aug 2025
• GPT Image 1 / 1.5 — OpenAI's Image Frontier — Apr 2025
• OpenAI o3 — Frontier Reasoning at Scale — Apr 2025
• OpenAI o1 — Inference-Time Reasoning — Sep 2024
• GPT-4o — Natively Multimodal OpenAI Frontier — May 2024
• Sora — OpenAI's Text-to-Video — Feb 2024
• GPT-4 — OpenAI's First Multimodal Frontier Model — Mar 2023
• ChatGPT — RLHF-Aligned Conversational LLM — Nov 2022
• Whisper — Open Multilingual Speech Recognition — Sep 2022
• InstructGPT — Training LMs to Follow Instructions with Human Feedback — Jan 2022
• CLIP — Contrastive Language-Image Pretraining — Jan 2021
• GPT-3 — Language Models are Few-Shot Learners — May 2020
• Scaling Laws for Neural Language Models (Kaplan et al.) — Jan 2020
• GPT-2 — Language Models are Unsupervised Multitask Learners — Feb 2019
• …and 2 more

Series

Benchmark snapshot

Best score within this company’s nodes (not a normalized cross-benchmark ranking).

• Speed · Output tok/s: 277 (from gpt-oss-20b)
• GPQA: 93.5% (42.8–93.5%) (from gpt-5-3)
• MMLU-Pro: 87.4% (55.6–87.4%) (from gpt-5-3)
• Agentic · Terminal-Bench Hard: 60.6% (6.8–60.6%) (from gpt-5-3)
• Coding · AA Coding Index: 59.10 (from gpt-5-3)
• Coding · SciCode: 56.6% (29.1–56.6%) (from gpt-5-3)
• Intelligence · AA Intelligence Index: 56.00 (from gpt-5-3)
• HLE: 44.3% (4.1–44.3%) (from gpt-5-3)

Official catalog (manual)

• gpt-5.1 — GPT‑5.1 (API) — in tree
• gpt-5.2 — GPT‑5.2 (Instant / Thinking / Pro) — in tree
• gpt-5.3-instant — GPT‑5.3 Instant — in tree
• gpt-5.3-codex — GPT‑5.3‑Codex
• gpt-5.3-codex-spark — GPT‑5.3‑Codex‑Spark (preview)
• gpt-5.4 — GPT‑5.4 (Thinking / Pro) — in tree
• gpt-5.4-mini — GPT‑5.4 mini
• gpt-5.4-nano — GPT‑5.4 nano
• gpt-5.5 — GPT‑5.5 (Thinking / Pro) — in tree
• gpt-5-4-mini — GPT-5.4 mini (xhigh) — in tree
• gpt-5-3-codex — GPT-5.3 Codex (xhigh) — in tree
• gpt-5-5-non-reasoning — GPT-5.5 (Non-reasoning) — in tree
• gpt-oss-120b — gpt-oss-120B (high) — in tree
• gpt-oss-120b-low — gpt-oss-120B (low) — in tree
• gpt-oss-20b-low — gpt-oss-20B (low) — in tree
• gpt-oss-20b — gpt-oss-20B (high) — in tree
• gpt-5-4-mini-medium — GPT-5.4 mini (medium) — in tree
• gpt-5-4-mini-non-reasoning — GPT-5.4 mini (Non-Reasoning) — in tree
• gpt-5-5 — GPT-5.5 (xhigh) — in tree
• gpt-5-4-nano — GPT-5.4 nano (xhigh) — in tree
• gpt-5-4-pro — GPT-5.4 Pro (xhigh) — in tree
• gpt-5-4-non-reasoning — GPT-5.4 (Non-reasoning) — in tree
• gpt-5-4-nano-medium — GPT-5.4 nano (medium) — in tree
• gpt-5-5-high — GPT-5.5 (high) — in tree
• gpt-5-5-low — GPT-5.5 (low) — in tree
• gpt-5-4-nano-non-reasoning — GPT-5.4 nano (Non-Reasoning) — in tree
• o3 — o3 — in tree
• gpt-5-5-medium — GPT-5.5 (medium) — in tree
• gpt-5-4 — GPT-5.4 (xhigh) — in tree
• o1 — o1 — in tree
• o1-preview — o1-preview — in tree
• o1-mini — o1-mini — in tree
• gpt-4o-2024-08-06 — GPT-4o (Aug '24) — in tree
• gpt-4o-2024-05-13 — GPT-4o (May '24) — in tree
• gpt-4-turbo — GPT-4 Turbo — in tree
• gpt-4o — GPT-4o (Nov '24) — in tree
• gpt-4o-mini — GPT-4o mini — in tree
• gpt-35-turbo — GPT-3.5 Turbo — in tree
• gpt-4o-realtime-dec-2024 — GPT-4o Realtime (Dec '24) — in tree
• gpt-5-1 — GPT-5.1 (high) — in tree
• o3-mini — o3-mini — in tree
• gpt-4 — GPT-4 — in tree
• gpt-5-minimal — GPT-5 (minimal) — in tree
• gpt-5-2-codex — GPT-5.2 Codex (xhigh) — in tree
• o4-mini — o4-mini (high) — in tree
• gpt-5-nano-minimal — GPT-5 nano (minimal) — in tree
• gpt-5-nano — GPT-5 nano (high) — in tree
• gpt-4-1-mini — GPT-4.1 mini — in tree
• gpt-4-1 — GPT-4.1 — in tree
• gpt-5-medium — GPT-5 (medium) — in tree
• o3-pro — o3-pro — in tree
• gpt-5 — GPT-5 (high) — in tree
• gpt-5-low — GPT-5 (low) — in tree
• gpt-5-nano-medium — GPT-5 nano (medium) — in tree
• gpt-4-1-nano — GPT-4.1 nano — in tree
• gpt-4-5 — GPT-4.5 (Preview) — in tree
• gpt-5-1-non-reasoning — GPT-5.1 (Non-reasoning) — in tree
• o3-mini-high — o3-mini (high) — in tree
• gpt-5-2-medium — GPT-5.2 (medium) — in tree
• gpt-5-2-non-reasoning — GPT-5.2 (Non-reasoning) — in tree
• gpt-5-chatgpt — GPT-5 (ChatGPT) — in tree
• gpt-5-2 — GPT-5.2 (xhigh) — in tree
• gpt-3-5-turbo-0613 — GPT-3.5 Turbo (0613) — in tree
• gpt-5-mini — GPT-5 mini (high) — in tree
• gpt-5-1-codex-mini — GPT-5.1 Codex mini (high) — in tree
• gpt-5-1-codex — GPT-5.1 Codex (high) — in tree
• gpt-4o-chatgpt-03-25 — GPT-4o (March 2025, chatgpt-4o-latest) — in tree
• gpt-5-mini-medium — GPT-5 mini (medium) — in tree
• gpt-4o-chatgpt — GPT-4o (ChatGPT) — in tree
• gpt-5-mini-minimal — GPT-5 mini (minimal) — in tree
• gpt-4o-mini-realtime-dec-2024 — GPT-4o mini Realtime (Dec '24) — in tree
• gpt-5-codex — GPT-5 Codex (high) — in tree
• o1-pro — o1-pro — in tree

Sources

• https://openai.com/
• https://openai.com/research/index/release/

Notes

This page is auto-assembled from nodes in the tree whose org matches one of the orgs above.