← AI Switchboard
AI Switchboard
AI Models · OpenAI · every model, newest first
← AI Model Directory

#1OpenAI

US AI lab that makes ChatGPT, Codex and the GPT model family, sold through apps and an API.

Top model GPT-6 AstraBased in San Francisco, USAopenai.com

Models10 releases · newest first

Latest29 Sep 2026Mid-size LLM

GPT-6.1 Sol

An upgrade to GPT-6 Sol that OpenAI says nearly matches GPT-6 Astra on agentic coding, computer use and professional work at one-fifth of Astra's standard token prices. It is the mid-priced model of the GPT-6 family and was presented at OpenAI's DevDay 2026.

  • 1,050,000-token context window, 128K max output
  • Text and image input, text output
  • Reasoning effort settings from low to max
  • Knowledge cutoff 30 April 2026

AccessChatGPT Work and Codex for Plus, Pro, Business, Enterprise and Edu users (not yet in regular ChatGPT chat at launch); OpenAI API as gpt-6.1-sol.

Price$2 input / $10 output per 1M tokens (prompts up to 272K tokens; higher rates above that)

Source: OpenAI - Introducing GPT-6.1 Sol · more · more
22 Sep 2026Mid-size LLM

GPT-6 Sol

The mid-tier model of the GPT-6 family, built with methods similar to GPT-6 Astra for complex coding and agentic workflows at lower cost. OpenAI priced it about 50% below its predecessor tier. A week later OpenAI released GPT-6.1 Sol as its upgrade.

  • 1,050,000-token context window, 128K max output
  • Text and image input, text output
  • Reasoning effort settings from none to max
  • Knowledge cutoff 20 April 2026

AccessChatGPT Work and Codex for Plus, Pro, Business, Enterprise and Edu users; OpenAI API as gpt-6-sol.

Price$2 input / $10 output per 1M tokens

Source: OpenAI - Introducing GPT-6 Sol and Luna · more · more
22 Sep 2026Small/fast LLM

GPT-6 Luna

OpenAI's most efficient GPT-6 model, aimed at focused, high-volume tasks. It carries the training advances behind GPT-6 Astra at a much lower price, and it is the GPT-6 model that Free and Go users can reach in the ChatGPT desktop app.

  • 1,050,000-token context window, 128K max output
  • Text and image input, text output
  • Reasoning effort settings from none to max
  • Knowledge cutoff 18 May 2026

AccessChatGPT Work and Codex for Plus, Pro, Business, Enterprise and Edu users; Free and Go users in the ChatGPT desktop app; OpenAI API as gpt-6-luna.

Price$0.10 input / $0.50 output per 1M tokens

Source: OpenAI - Introducing GPT-6 Sol and Luna · more · more
8 Sep 2026Image model

GPT Image 2.5 Sunburst and GPT Image 2.5 Flare (ChatGPT Images 2.5)

OpenAI's image generation and editing models, offered in ChatGPT as Images 2.5. Flare is the default for most applications, which OpenAI says beats GPT Image 2 on quality at 50% lower latency; Sunburst is aimed at premium work that needs tighter control across edits.

  • Text and image input, image output
  • Quality settings from low up to xhigh and max
  • Up to 50% lower generation latency than Images 2.0 (OpenAI)
  • Dated API snapshots 2026-09-08

AccessAll ChatGPT, ChatGPT Work and Codex users on desktop, mobile and web (as Images 2.5); OpenAI API as gpt-image-2.5-sunburst and gpt-image-2.5-flare.

Price$5 per 1M text input tokens / $30 per 1M image output tokens (standard tier, both models)

Source: OpenAI - Introducing ChatGPT Images 2.5 · more · more
3 Sep 2026Frontier LLM

GPT-6 Astra

The first and most capable model of OpenAI's GPT-6 family, which OpenAI positions as its best model across the board, built for the hardest end-to-end work. OpenAI highlights computer use, browsing, software engineering, cybersecurity, science and professional work. An Ultrafast mode was added on 29 September 2026.

  • 1,050,000-token context window, 128K max output
  • Text and image input, text output
  • Reasoning effort settings from low to max
  • Knowledge cutoff 30 April 2026

AccessRolled out first to a limited set of organizations, then to ChatGPT Plus, Pro, Business and Enterprise; OpenAI API as gpt-6-astra; Microsoft Azure; AWS Bedrock; Codex.

Price$10 input / $50 output per 1M tokens (standard)

Source: OpenAI - GPT-6 Astra: A new generation of intelligence · more · more
9 Jul 2026Frontier LLM

GPT-5.6 (Sol, Terra and Luna)

A three-model family that introduced OpenAI's Sol, Terra and Luna tiers: Sol as the flagship for coding, knowledge work, cybersecurity and science, Terra as a balanced model OpenAI calls competitive with GPT-5.5, and Luna as the most cost-efficient. OpenAI cut its prices twice after launch.

  • Sol: 1,050,000-token context window, 128K max output
  • Programmatic tool calling
  • Text and image input, text output
  • Sol knowledge cutoff 16 February 2026

AccessChatGPT (Plus, Pro, Business and Enterprise use Sol at medium and higher effort); Codex (Free and Go get Terra, paid plans get all three); OpenAI API for all three.

PriceAt launch: Sol $5/$30, Terra $2.50/$15, Luna $1/$6 per 1M input/output tokens; cut on Jul 30 and Aug 21 (docs now list Sol at $4/$20, promotional to 21 Nov 2026)

Source: OpenAI - GPT-5.6: Frontier intelligence that scales with your ambition · more · more
23 Apr 2026Frontier LLM

GPT-5.5

A frontier model that OpenAI called a new class of intelligence for real work, aimed at software development, research and data analysis. OpenAI said it matches GPT-5.4's per-token latency in real-world serving while performing at a much higher level. A Pro variant uses more compute.

  • 1,050,000-token context window in the API, 128K max output
  • 400K context window in Codex
  • Text and image input, text output
  • Knowledge cutoff 1 December 2025

AccessChatGPT Plus, Pro, Business and Enterprise; GPT-5.5 Pro for Pro, Business and Enterprise; Codex; OpenAI API from 24 April 2026.

Price$5 input / $30 output per 1M tokens; GPT-5.5 Pro $30 / $180

Source: OpenAI - Introducing GPT-5.5 · more · more
5 Mar 2026Frontier LLM

GPT-5.4

OpenAI's frontier model for professional work as of March 2026, combining its reasoning, coding and agentic advances in a single model. It added native computer-use abilities so agents can operate computers across applications. Smaller GPT-5.4 mini and nano models followed on 17 March.

  • Native computer use for agents
  • 1,050,000-token context window in the API, 128K max output
  • Text and image input, text output
  • Mini and nano sizes released 17 March 2026

AccessChatGPT (GPT-5.4 Thinking for Plus, Team and Pro; GPT-5.4 Pro for Pro and Enterprise); Codex; OpenAI API as gpt-5.4 and gpt-5.4-pro.

Price$2.50 input / $15 output per 1M tokens; GPT-5.4 Pro $30 / $180

Source: OpenAI - Introducing GPT-5.4 · more · more
5 Feb 2026Coding model

GPT-5.3-Codex

An agentic coding model that OpenAI said pairs frontier coding performance with general reasoning for long-running technical work, from debugging to deployment and monitoring. OpenAI reported it runs 25% faster than its predecessor. It has since been marked deprecated in OpenAI's API documentation.

  • SWE-Bench Pro 56.8% (OpenAI-reported)
  • Terminal-Bench 2.0 77.3% (OpenAI-reported)
  • 400K-token context window, 128K max output
  • Text and image input, text output

AccessCodex app, CLI, IDE extension and web on paid ChatGPT plans; API access came later. The API model page now marks it deprecated.

Price$1.75 input / $14 output per 1M tokens (API model page)

Source: OpenAI - Introducing GPT-5.3-Codex · more
11 Dec 2025Frontier LLM

GPT-5.2

OpenAI's flagship for professional knowledge work from December 2025, with gains in spreadsheets, presentations, coding, image analysis, long-context work and tool use. It shipped as Instant, Thinking and Pro variants. OpenAI's API docs now call it a previous flagship and point new projects to GPT-6 Astra.

  • Instant, Thinking and Pro variants
  • 400K-token context window, 128K max output (API docs)
  • Text and image input, text output
  • Knowledge cutoff 31 August 2025

AccessChatGPT paid plans (Plus, Pro, Go, Business, Enterprise) as Instant, Thinking and Pro; OpenAI API the same day.

Price$1.75 input / $14 output per 1M tokens; GPT-5.2 Pro $21 / $168

Source: OpenAI - Introducing GPT-5.2 · more · more

What the industry says · 37 voices since 8 Apr 2026 · rank #1

  • Nathan Benaich, Air Street Capital, 8 Oct 2026: puts OpenAI at the frontier. “The frontier is now a three-lab race between Anthropic, OpenAI, and Google.” source
  • Artificial Analysis, 8 Oct 2026: puts OpenAI at the frontier (cyber only). “GPT-6 Sol (Daybreak Blue, max) now leads the Index and sits on the Cost vs. Capability Pareto frontier.” source
  • Jörn Brien, t3n, 6 Oct 2026: puts OpenAI at the frontier. “Anthropics Claude Opus 5.5, OpenAIs GPT-6 Astra oder Googles Gemini-4 Argon sind mit jeweils über 50 Punkten zwar weiter deutlich voraus, sind aber auch nicht Open-Source.” source
  • J.D. Capelouto, Semafor, 5 Oct 2026: puts OpenAI at the frontier. “Alibaba's most advanced model, Qwen 3.8-Max, which trails the top closed American models from OpenAI and Anthropic.” source
  • Andrew Sharp, Stratechery, 2 Oct 2026: puts OpenAI at the frontier. “It was only a few weeks ago that it seemed like OpenAI had figured it out, was the clear leader on the frontier, and had settled on a coherent strategic direction.” source
  • JPMorgan equity research, 1 Oct 2026: names OpenAI among the leaders. source
  • Nick Patience, Futurum, 1 Oct 2026: puts OpenAI at or near the frontier. “On the early independent evidence, Argon is level with OpenAI's best model and behind Anthropic's, which is a strong recovery for a company that went more than seven months without a new frontier model, yet short of the lead that Google's own benchmark table implies.” source
  • Matthias Bastian, The Decoder, 1 Oct 2026: puts OpenAI at the frontier. “closes the gap with frontier models from OpenAI and Anthropic, beating some of them on key benchmarks” source
29 more
  • Polymarket via DeFi Rate, 26 Sep 2026: ranked OpenAI #3 of 15. source
  • Zvi Mowshowitz, 26 Sep 2026: ranked OpenAI #2 of 2. source
  • Elon Musk, SpaceX, 25 Sep 2026: puts OpenAI at the frontier. “I am cautiously optimistic that SpaceX will have a Fable/GPT-6 level model in 2 to 3 months” source
  • Kalshi via Covers, 24 Sep 2026: ranked OpenAI #2 of 7. “Best AI 2026 Odds: Claude Extends Prediction Market Lead, OpenAI Slumps After Latest Hack” source
  • Nathan Lambert, Interconnects (Ai2), 21 Sep 2026: puts OpenAI at the frontier. “Together, Chinese open-weight models are approximately 2-5 months behind the closed American frontier, with the open-weight American models being approximately 6-9 months behind the likes of OpenAI and Anthropic.” source
  • Ben Thompson, Stratechery, 21 Sep 2026: puts OpenAI at the frontier. “I'm not sure it's a coincidence the demand to pace the frontier came when the frontier was, for the first time in a while, set by OpenAI.” source
  • Menlo Ventures, 16 Sep 2026: ranked OpenAI #1 of 3. “Among AI users, ChatGPT remains the most widely used assistant at 60%, up from 45% last year, but its lead is narrowing. Gemini grew from 36% to 58% in 2026, cutting ChatGPT's lead from nine points to just two. ... Claude grew fastest, with usage nearly tripling from 7% in 2025 to 20% in 2026.” source
  • Epoch AI, 9 Sep 2026: ranked OpenAI #1 of 5. “By our median estimates, OpenAI may have had a narrow lead over Google DeepMind at the end of last year, though the uncertainty intervals for both substantially overlap. ... Meanwhile, Anthropic was almost certainly well behind OpenAI in compute in 2025, and SpaceXAI was probably fifth.” source
  • Gartner, 1 Sep 2026: names OpenAI among the leaders. source
  • Jensen Huang, NVIDIA, 26 Aug 2026: names OpenAI among the leaders. “NVIDIA runs the leading closed models; OpenAI, Anthropic, Grok, Meta, Gemini, and the leading open models; TML, Mistral, Qwen, Kimi, GLM, DeepSeek, MiniMax, and Nemotron.” source
  • Ian Krietzberg, Puck, 20 Aug 2026: puts OpenAI at the frontier. “Until earlier this month, OpenAI and Anthropic were considered the undisputed frontrunners of the A.I. frontier race—having amassed a seemingly insurmountable lead, in terms of both dollars and mindshare, over developers such as SpaceXAI and Meta.” source
  • Ben Wood, CCS Insight, 12 Aug 2026: puts OpenAI at the frontier. “The goal will undoubtedly be to close the gaps with Anthropic and OpenAI in some areas.” source
  • Beatrice Nolan, Fortune, 5 Aug 2026: puts OpenAI at the frontier. “aren't currently competitive with the bleeding-edge American AI systems from Anthropic, OpenAI, or Google DeepMind” source
  • Ethan Mollick, Wharton, 23 Jul 2026: puts OpenAI at the frontier. “Google, which led on benchmarks not that long ago, has fallen behind where it now counts: it has no leading frontier model and it has nothing close to Codex and Code.” source
  • Alex Kantrowitz, Big Technology, 17 Jul 2026: puts OpenAI at the frontier. source
  • Raffi Krikorian, Mozilla, 16 Jul 2026: names OpenAI among the leaders. source
  • SemiAnalysis, 9 Jul 2026: puts OpenAI at the frontier. “Since then, frontier AI has increasingly felt like a two horse race between OpenAI vs Anthropic.” source
  • M.G. Siegler, Spyglass, 23 Jun 2026: puts OpenAI at the frontier. “Or perhaps it's an acknowledgement that no matter the work they may do at the frontier, they're not going to be able to catch Anthropic and OpenAI at this point.” source
  • Bob Gourley, OODA Loop, 16 Jun 2026: names OpenAI among the leaders. source
  • Alexandr Wang, Meta, 4 Jun 2026: puts OpenAI at the frontier. source
  • Satya Nadella, Microsoft, 4 Jun 2026: names OpenAI among the leaders. “there is OpenAI, there is Anthropic, there's Google, there is tons of folks who are in there” source
  • CNBC, 23 May 2026: ranked OpenAI #2 of 50 (private companies only). source
  • 5W AI Visibility Index, 13 May 2026: ranked OpenAI #1 of 25. “In the dataset, the most-cited AI company — OpenAI — accounts for 24.6% of cited responses across the five engines.” source
  • Ramp AI Index, 13 May 2026: ranked OpenAI #2 of 2. “Today, for the first time, Anthropic passed OpenAI in business adoption ... Adoption of Anthropic rose 3.8% in April to 34.4% of businesses. ... OpenAI adoption fell 2.9% to 32.3%.” source
  • NIST's Center for AI Standards and Innovation (CAISI), 1 May 2026: names OpenAI among the leaders. source
  • The AI Daily Brief (Nathaniel Whittemore), 29 Apr 2026: ranked OpenAI #1 of 8. “I had OpenAI and Google tied at 74, Anthropic at 70, and the rest clustered between Apple at 58 and Amazon at 64.” source
  • Gene Munster, Deepwater Asset Management, 28 Apr 2026: ranked OpenAI #1 of 2 (coding only). “If you're serious about developing, you're using GPT-5, you're using Codex. ... You're not using Anthropic.” source
  • Stanford AI Index, 13 Apr 2026: ranked OpenAI #1 of 3. “Within industry, a small set of organizations account for a large share of releases. In 2025, the top contributors were OpenAI (20), Google (14), and Alibaba (11).” source
  • The Next Web, 8 Apr 2026: ranked OpenAI #1 of 4. “ranked fourth on the Artificial Analysis Intelligence Index v4.0, with a score of 52, behind Gemini 3.1 Pro Preview and GPT-5.4 (both at 57) and Claude Opus 4.6 (53)” source
← All labs#2 Anthropic →