Skip to main content

BenchLM recommendation

Best OpenAI Models in 2026

Data verified

As of July 20, 2026, the top model in best openai models on the BenchLM leaderboard is GPT-5.6 Sol with a score of 82.

Last verified: July 20, 2026

All OpenAI models ranked by benchmark performance — GPT-5, GPT-4o, o1, o3, and more.

Unless noted otherwise, ranking surfaces on this page use BenchLM's provisional leaderboard lane rather than the stricter sourced-only verified leaderboard.

Bottom line: GPT-5.4 is OpenAI's strongest model — it leads on knowledge and holds top 3 overall. GPT-5.4 Pro adds premium reasoning at higher cost. GPT-5.3 Codex is the coding specialist.

GPT-5.6 Sol leads this ranking with a score of 82, followed by GPT-5.4 (74.2) and GPT-5.5 (73.5). There is a significant gap between the leading models and the rest of the field.

The best open-weight option is GPT-OSS 120B (ranked #23 with a score of 50.1). Proprietary models hold a clear advantage in this category, though open-weight options may suffice for less demanding use cases.

This ranking is based on provisional overall weighted scores across BenchLM.ai's scoring formula tracked by BenchLM.ai. For detailed model profiles, click any model name below. To compare two specific models head-to-head, use the "vs #" links.

What changed

GPT-5.4 leads OpenAI's lineup with the highest overall score and best knowledge (98).

GPT-5.4 Pro premium tier with perfect multimodal and math scores.

GPT-5.3 Codex coding-focused with perfect math and multilingual.

How to choose

Full Rankings (38 models)

1
GPT-5.6 Sol
OpenAI·Proprietary·1M

82

BenchAlign v5

2
GPT-5.4
OpenAI·Proprietary·1.05M

74.2

BenchAlign v5

3
GPT-5.5
OpenAI·Proprietary·1M

73.5

BenchAlign v5

4
GPT-5.6 Terra
OpenAI·Proprietary·1M

72.6

BenchAlign v5

5
GPT-5.6 Luna
OpenAI·Proprietary·1M

67.2

BenchAlign v5

6
GPT-5.2 Pro
OpenAI·Proprietary·400K

67

BenchAlign v5

7
GPT-5.4 nano
OpenAI·Proprietary·400K

66.8

BenchAlign v5

8
GPT-5.3 Codex
OpenAI·Proprietary·400K

66.7

BenchAlign v5

9
GPT-5.5 Pro
OpenAI·Proprietary·1M

63.7

BenchAlign v5

10
GPT-5.4 Pro
OpenAI·Proprietary·1.05M

60.9

BenchAlign v5

11
GPT-5.2-Codex
OpenAI·Proprietary·400K

59.1

BenchAlign v5

12
GPT-5.2 Instant
OpenAI·Proprietary·128K

59

BenchAlign v5

13
GPT-5.3 Instant
OpenAI·Proprietary·400K

58.9

BenchAlign v5

14
GPT-5 (high)
OpenAI·Proprietary·128K

58.6

BenchAlign v5

15
GPT-5.2
OpenAI·Proprietary·400K

58.4

BenchAlign v5

16
GPT-5.3-Codex-Spark
OpenAI·Proprietary·256K

56.9

BenchAlign v5

17
GPT-5.4 mini
OpenAI·Proprietary·400K

56.8

BenchAlign v5

18
GPT-5 (medium)
OpenAI·Proprietary·128K

55.2

BenchAlign v5

19
GPT-5.1-Codex-Max
OpenAI·Proprietary·400K

54.5

BenchAlign v5

20
GPT-5.1
OpenAI·Proprietary·200K

53.7

BenchAlign v5

21
GPT-5.1-Codex
OpenAI·Proprietary·400K

52.7

BenchAlign v5

22
GPT-4.1
OpenAI·Proprietary·1M

51.1

BenchAlign v5

23
GPT-OSS 120B
OpenAI·Open Weight·128K

50.1

BenchAlign v5

24
o4-mini (high)
OpenAI·Proprietary·200K

50

BenchAlign v5

25
o1-preview
OpenAI·Proprietary·200K

49.1

BenchAlign v5

26
o3-pro
OpenAI·Proprietary·200K

48.3

BenchAlign v5

27
o1
OpenAI·Proprietary·200K

48.1

BenchAlign v5

28
o3
OpenAI·Proprietary·200K

47.9

BenchAlign v5

29
o3-mini
OpenAI·Proprietary·200K

47.4

BenchAlign v5

30
GPT-5 nano
OpenAI·Proprietary·400K

46.4

BenchAlign v5

31
o1-pro
OpenAI·Proprietary·200K

45.9

BenchAlign v5

32
GPT-4.1 mini
OpenAI·Proprietary·1M

44.2

BenchAlign v5

33
GPT-5 mini
OpenAI·Proprietary·128K

43.9

BenchAlign v5

34
GPT-OSS 20B
OpenAI·Open Weight·128K

42.7

BenchAlign v5

35
GPT-4.1 nano
OpenAI·Proprietary·1M

42.1

BenchAlign v5

36
GPT-4o
OpenAI·Proprietary·128K

41.5

BenchAlign v5

37
GPT-4o mini
OpenAI·Proprietary·128K

37.9

BenchAlign v5

38
GPT-4 Turbo
OpenAI·Proprietary·128K

27.4

BenchAlign v5

Key Takeaways

The top model is GPT-5.6 Sol by OpenAI with a BenchAlign v5 score of 82 and Supported evidence.

The best open-weight model is GPT-OSS 120B at position #23.

38 models are included in this ranking.

Score in Context

What these scores mean

Models are ranked by the same overall BenchLM score used across all leaderboards. Comparing within OpenAI's lineup helps identify which model fits your use case and budget.

Known limitations

This page only shows OpenAI models. Cross-provider comparison requires the overall or category-specific leaderboards. Newer models may have limited benchmark coverage initially.

Last updated: July 20, 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.