Benchmark Almanac

OpenAI

Best on the Capabilities Index GPT-6 Astra 169.1
90 models 48 model cards read openai.com

OpenAI's models against the field

Every model on the Epoch Capabilities Index by release date, with OpenAI's in color.

50 75 100 125 150 175 2023 2024 2025 2026 2027 GPT-6.1 Terra GPT-5.6 Sol 50 75 100 125 150 175 2023 2024 2025 2026 2027 GPT-6.1 Terra GPT-5.4 Pro

Its strongest models

OpenAI Sep 2026

GPT-6 Astra

Closed weights Flagship

169.1 #1 of 271
Epoch Capabilities Index
  • Humanity's Last Exam 54.8% #1 /47
  • DeepSWE 74.1% #1 /28
  • AutomationBench 41.4% #5 /5
OpenAI Oct 2026

GPT-6.1 Terra

Closed weights

168.1 #3 of 271
Epoch Capabilities Index
  • Agents' Last Exam 41.8% #5 /12
  • MultiChallenge 74.2% #7 /35
  • HealthBench Professional 62.7% #9 /25
OpenAI Sep 2026

GPT-6.1 Sol

Closed weights Mid-size

167.8 #4 of 271
Epoch Capabilities Index
  • Terminal-Bench 4.0 58.2% #4 /23
  • MultiChallenge 82% #1 /35
  • HealthBench Professional 64.2% #8 /25
OpenAI Sep 2026

GPT-6 Sol

Closed weights Mid-size

163.8 #9 of 271
Epoch Capabilities Index
  • Terminal-Bench 4.0 49.4% #7 /23
  • ZeroBench 41% #3 /66
  • HealthBench Professional 60.8% #13 /25
OpenAI Apr 2026

GPT-5.5 Pro

Closed weights

163.1 #10 of 271
Epoch Capabilities Index
  • CritPt 30.6% #9 /160
  • GPQA Diamond 93.9% #14 /219
  • FrontierMath 52.4% #1 /70
OpenAI Jul 2026

GPT-5.6 Sol

Closed weights Flagship

162.6 #12 of 271
Epoch Capabilities Index
  • DeepSWE 72.7% #4 /28
  • Agents' Last Exam 30.3% #9 /12
  • Terminal-Bench 4.0 37.3% #11 /23

What its model cards measure

Tall ticks are OpenAI model cards that list the benchmark in their headline table. Gray dots are cards that leave it out. Rows are the benchmarks the lab listed most, at least twice each, in the order it started using them.

MMLU GPQA Diamond MGSM MMMU MathVista AIME 2024 Codeforces rating SWE-bench Verified tau-bench FrontierMath Humanity's Last Exam SWE-Lancer CharXiv Aider Polyglot MultiChallenge GraphWalks BrowseComp AIME 2025 MMMU-Pro tau2-bench Terminal-Bench 2.0 SWE-Bench Pro GDPval Toolathlon Agents' Last Exam DeepSWE OSWorld 2.0 AutomationBench

Model cards

DateModelDocumentBenchmarks
26 Oct 2026 GPT-6.1 Terra Introducing GPT-6.1 Terra (OpenAI) 5
29 Sep 2026 GPT-6.1 Sol Introducing GPT-6.1 Sol 5
22 Sep 2026 GPT-6 Luna Introducing GPT-6 Sol and Luna (OpenAI) 5
22 Sep 2026 GPT-6 Sol Introducing GPT-6 Sol and Luna 5
3 Sep 2026 GPT-6 Astra GPT-6 Astra: A new generation of intelligence 18
9 Jul 2026 GPT-5.6 Luna GPT-5.6 (OpenAI) 27
9 Jul 2026 GPT-5.6 Terra GPT-5.6 (OpenAI) 27
9 Jul 2026 GPT-5.6 Sol GPT-5.6: Frontier intelligence that scales with your ambition 11
7 May 2026 GPT-Realtime-2 Advancing voice intelligence with new models in the API (OpenAI) 2
5 May 2026 GPT-5.5 Instant GPT-5.5 Instant (OpenAI) 5
23 Apr 2026 GPT-5.5 Pro Introducing GPT-5.5 (OpenAI) 7
23 Apr 2026 GPT-5.5 Introducing GPT-5.5 11
17 Mar 2026 GPT-5.4 Nano Introducing GPT-5.4 mini and nano (OpenAI) 12
17 Mar 2026 GPT-5.4 Mini Introducing GPT-5.4 mini and nano 5
5 Mar 2026 GPT-5.4 Pro Introducing GPT-5.4 (OpenAI) 11
5 Mar 2026 GPT-5.4 Introducing GPT-5.4 8
12 Feb 2026 GPT-5.3-Codex-Spark Introducing GPT-5.3-Codex-Spark (OpenAI) 2
5 Feb 2026 GPT-5.3 Codex Introducing GPT-5.3-Codex 5
18 Dec 2025 GPT-5.2 Codex Introducing GPT-5.2-Codex (OpenAI) 3
11 Dec 2025 GPT-5.2 Pro Introducing GPT-5.2 (OpenAI) 9
11 Dec 2025 GPT-5.2 Introducing GPT-5.2 13
19 Nov 2025 GPT-5.1-Codex-Max Building more with GPT-5.1-Codex-Max (OpenAI) 3
13 Nov 2025 GPT-5.1 Introducing GPT-5.1 for developers (Appendix: Model evaluations) 7
15 Sep 2025 GPT-5-Codex Introducing upgrades to Codex (OpenAI) 2
28 Aug 2025 gpt-realtime Introducing gpt-realtime and Realtime API updates for production voice agents (OpenAI) 3
7 Aug 2025 GPT-5 nano Introducing GPT-5 for developers (OpenAI) 23
7 Aug 2025 GPT-5 mini Introducing GPT-5 for developers (OpenAI) 23
7 Aug 2025 GPT-5 Introducing GPT-5 21
5 Aug 2025 gpt-oss-20b Introducing gpt-oss (OpenAI) 9
5 Aug 2025 gpt-oss-120b Introducing gpt-oss 9
17 Jul 2025 ChatGPT agent Introducing ChatGPT agent: bridging research and action (OpenAI) 7
10 Jun 2025 o3-pro Launching OpenAI o3-pro (OpenAI Model Release Notes) 3
16 May 2025 codex-1 Introducing Codex (OpenAI) 1
16 Apr 2025 o4-mini Introducing OpenAI o3 and o4-mini (OpenAI) 14
16 Apr 2025 o3 Introducing OpenAI o3 and o4-mini 14
14 Apr 2025 GPT-4.1 nano Introducing GPT-4.1 in the API (OpenAI) 18
14 Apr 2025 GPT-4.1 mini Introducing GPT-4.1 in the API (OpenAI) 19
14 Apr 2025 GPT-4.1 Introducing GPT-4.1 in the API 10
27 Feb 2025 GPT-4.5 Introducing GPT-4.5 8
2 Feb 2025 OpenAI deep research Introducing deep research (OpenAI) 2
31 Jan 2025 o3-mini OpenAI o3-mini 10
23 Jan 2025 OpenAI Computer-Using Agent (CUA) Computer-Using Agent (OpenAI) 3
17 Dec 2024 o1 OpenAI o1 and new tools for developers 11
12 Sep 2024 o1-preview Learning to Reason with LLMs (Appendix A) 7
18 Jul 2024 GPT-4o mini GPT-4o mini: advancing cost-efficient intelligence 8
13 May 2024 GPT-4o May 2024 Hello GPT-4o 14
11 Apr 2024 GPT-4 Turbo Apr 2024 openai/simple-evals README (initial results table) 6
14 Mar 2023 GPT-4 Mar 2023 GPT-4 Technical Report (Table 2) + GPT-4 research post (visual inputs table) 15

Every OpenAI model