Skip to main content

Best Proprietary LLMs in 2026

Top proprietary/closed-source AI models ranked by benchmark performance.

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

Bottom line: Claude Mythos Preview leads proprietary models overall, with Gemini 3.1 Pro and GPT-5.4 close behind. The top tier is competitive — choice often comes down to pricing, latency, and API ecosystem.

According to BenchLM.ai, Claude Mythos Preview leads this ranking with a score of 99, followed by Claude Opus 4.7 (97) and GPT-5.4 (93). There is meaningful separation between the top models, suggesting genuine performance differences.

All models in this ranking are proprietary. No open-weight alternatives are available for this category.

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

Claude Mythos Preview entered at #1 across proprietary models with the highest overall score.

Gemini 3.1 Pro strong #2 among proprietary models — best non-reasoning option.

GPT-5.4 leads knowledge benchmarks at 98 and holds a strong #3.

How to choose

Full Rankings (61 models)

Claude Mythos Preview
Anthropic·Proprietary·1M

99

prov. overall

Claude Opus 4.7
Anthropic·Proprietary·1M

97

prov. overall

GPT-5.4
OpenAI·Proprietary·1.05M

93

prov. overall

4
Gemini 3.1 Pro
Google·Proprietary·1M

93

prov. overall

5
GPT-5.4 Pro
OpenAI·Proprietary·1.05M

92

prov. overall

6
Claude Opus 4.6
Anthropic·Proprietary·1M

91

prov. overall

7
GPT-5.3 Codex
OpenAI·Proprietary·400K

89

prov. overall

8
Gemini 3 Pro Deep Think
Google·Proprietary·2M

86

prov. overall

9
Claude Sonnet 4.6
Anthropic·Proprietary·200K

85

prov. overall

10
GPT-5.2
OpenAI·Proprietary·400K

83

prov. overall

11
Gemini 3 Pro
Google·Proprietary·2M

83

prov. overall

12
Claude Opus 4.5
Anthropic·Proprietary·200K

80

prov. overall

13
GPT-5.1
OpenAI·Proprietary·200K

80

prov. overall

14
GPT-5.2-Codex
OpenAI·Proprietary·400K

80

prov. overall

15
Grok 4.1
xAI·Proprietary·1M

80

prov. overall

16
GPT-5 (high)
OpenAI·Proprietary·128K

79

prov. overall

17
Kimi K2.5 (Reasoning)
Moonshot AI·Proprietary·128K

79

prov. overall

18
GPT-5.1-Codex-Max
OpenAI·Proprietary·400K

78

prov. overall

19
Qwen3.6 Plus
Alibaba·Proprietary·1M

77

prov. overall

20
Grok 4.20
xAI·Proprietary·2M

77

prov. overall

21
GPT-5.4 mini
OpenAI·Proprietary·400K

73

prov. overall

22
GPT-5 (medium)
OpenAI·Proprietary·128K

73

prov. overall

23
Grok 4.1 Fast
xAI·Proprietary·1M

72

prov. overall

24
Claude Sonnet 4.5
Anthropic·Proprietary·200K

68

prov. overall

25
o1-preview
OpenAI·Proprietary·200K

68

prov. overall

26
Gemini 2.5 Pro
Google·Proprietary·1M

67

prov. overall

27
Gemini 3 Flash
Google·Proprietary·1M

67

prov. overall

28
Grok 4
xAI·Proprietary·128K

67

prov. overall

29
GPT-4.1
OpenAI·Proprietary·1M

60

prov. overall

30
Claude Haiku 4.5
Anthropic·Proprietary·200K

59

prov. overall

31
o1
OpenAI·Proprietary·200K

59

prov. overall

32
o3
OpenAI·Proprietary·200K

59

prov. overall

33
o3-pro
OpenAI·Proprietary·200K

59

prov. overall

34
o3-mini
OpenAI·Proprietary·200K

58

prov. overall

35
Claude 4.1 Opus
Anthropic·Proprietary·200K

53

prov. overall

36
Claude 4 Sonnet
Anthropic·Proprietary·200K

52

prov. overall

37
Mistral Large 3
Mistral·Proprietary·128K

52

prov. overall

38
Gemini 3.1 Flash-Lite
Google·Proprietary·1M

51

prov. overall

39
GPT-4.1 mini
OpenAI·Proprietary·1M

47

prov. overall

40
o4-mini (high)
OpenAI·Proprietary·200K

46

prov. overall

41
Claude 4.1 Opus Thinking
Anthropic·Proprietary·200K

45

prov. overall

42
GPT-4o mini
OpenAI·Proprietary·128K

45

prov. overall

43
Kimi K2
Moonshot AI·Proprietary·128K

43

prov. overall

44
Claude 3.5 Sonnet
Anthropic·Proprietary·200K

42

prov. overall

45
Grok Code Fast 1
xAI·Proprietary·256K

42

prov. overall

46
GPT-4o
OpenAI·Proprietary·128K

41

prov. overall

47
Gemini 2.5 Flash
Google·Proprietary·1M

40

prov. overall

48
Mistral Large 2
Mistral·Proprietary·128K

40

prov. overall

49
Gemini 1.5 Pro
Google·Proprietary·2M

37

prov. overall

50
Claude 3 Opus
Anthropic·Proprietary·200K

36

prov. overall

51
Grok 3 [Beta]
xAI·Proprietary·128K

34

prov. overall

52
o1-pro
OpenAI·Proprietary·200K

30

prov. overall

53
GLM-4.5
Z.AI·Proprietary·128K

29

prov. overall

54
GPT-4.1 nano
OpenAI·Proprietary·1M

28

prov. overall

55
GPT-4 Turbo
OpenAI·Proprietary·128K

27

prov. overall

56
Gemini 1.0 Pro
Google·Proprietary·32K

25

prov. overall

57
Z-1
Z·Proprietary·128K

25

prov. overall

58
Claude 3 Haiku
Anthropic·Proprietary·200K

24

prov. overall

59
Moonshot v1
Moonshot AI·Proprietary·128K

24

prov. overall

60
GLM-4.5-Air
Z.AI·Proprietary·128K

21

prov. overall

61
Nova Pro
Amazon·Proprietary·128K

11

prov. overall

These rankings update weekly

Get notified when models move. One email a week with what changed and why.

Free. No spam. Unsubscribe anytime.

Key Takeaways

The top model is Claude Mythos Preview by Anthropic with a provisional score of 99.

61 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 proprietary's lineup helps identify which model fits your use case and budget.

Known limitations

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

Last updated: April 20, 2026

The AI models change fast. We track them for you.

For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.