| A | B | C | D | E | F | G | H | I | J | K | |
|---|---|---|---|---|---|---|---|---|---|---|---|
1 | Count | ⬝ ▪ ◾ ⬛ | Model name | Params | Special | MMLU | Date | Full string | Source | ||
2 | ---- | ||||||||||
3 | 1. | ⬝ | GPT-1 | 117M | Jun/2018 | 1. ⬝ GPT-1 117M Jun/2018 | |||||
4 | ---- | GPT-2 | |||||||||
5 | 2. | ⬝ | GPT-2-Small | 124M | Feb/2019 | 2. ⬝ GPT-2-Small 124M Feb/2019 | |||||
6 | 3. | ⬝ | GPT-2-Medium | 355M | 3. ⬝ GPT-2-Medium 355M | ||||||
7 | 4. | ⬝ | GPT-2-Large | 774M | 4. ⬝ GPT-2-Large 774M | ||||||
8 | 5. | ⬝ | GPT-2-XL | 1.5B | [32.4] | 5. ⬝ GPT-2-XL 1.5B [32.4] | |||||
9 | ---- | GPT-3 | |||||||||
10 | 6. | ⬝ | GPT-3 Small 125M | 6. ⬝ GPT-3 Small 125M | |||||||
11 | 7. | ⬝ | GPT-3 Large 760M | 7. ⬝ GPT-3 Large 760M | |||||||
12 | 8. | ⬝ | GPT-3 2.7B | 8. ⬝ GPT-3 2.7B | |||||||
13 | 9. | ⬝ | GPT-3 13B | 9. ⬝ GPT-3 13B | |||||||
14 | 10. | ⬝ | GPT-3 ada | 350M | 10. ⬝ GPT-3 ada 350M | ||||||
15 | 11. | ⬝ | GPT-3 babbage | 1.3B | 11. ⬝ GPT-3 babbage 1.3B | ||||||
16 | 12. | ⬝ | GPT-3 curie | 6.7B | 12. ⬝ GPT-3 curie 6.7B | ||||||
17 | 13. | ▪ | GPT-3 davinci | 175B | [43.9] | May/2020 | 13. ▪ GPT-3 davinci 175B [43.9] May/2020 | ||||
18 | 14. | ⬝ | cushman:2020-05-03 | 14. ⬝ cushman:2020-05-03 | |||||||
19 | 15. | ⬝ | ada:2020-05-03 | 15. ⬝ ada:2020-05-03 | |||||||
20 | 16. | ⬝ | babbage:2020-05-03 | 16. ⬝ babbage:2020-05-03 | |||||||
21 | 17. | ⬝ | curie:2020-05-03 | 17. ⬝ curie:2020-05-03 | |||||||
22 | 18. | ▪ | davinci:2020-05-03 | 18. ▪ davinci:2020-05-03 | |||||||
23 | ---- | Special | |||||||||
24 | 19. | ⬝ | text-embedding-ada-002 | 19. ⬝ text-embedding-ada-002 | |||||||
25 | 20. | ⬝ | text-similarity-ada-001 | 20. ⬝ text-similarity-ada-001 | |||||||
26 | 21. | ⬝ | text-similarity-babbage-001 | 21. ⬝ text-similarity-babbage-001 | |||||||
27 | 22. | ⬝ | text-similarity-curie-001 | 22. ⬝ text-similarity-curie-001 | |||||||
28 | 23. | ▪ | text-similarity-davinci-001 | 23. ▪ text-similarity-davinci-001 | |||||||
29 | 24. | ⬝ | text-search-ada-doc-001 | 24. ⬝ text-search-ada-doc-001 | |||||||
30 | 25. | ⬝ | text-search-ada-query-001 | 25. ⬝ text-search-ada-query-001 | |||||||
31 | 26. | ⬝ | text-search-babbage-doc-001 | 26. ⬝ text-search-babbage-doc-001 | |||||||
32 | 27. | ⬝ | text-search-babbage-query-001 | 27. ⬝ text-search-babbage-query-001 | |||||||
33 | 28. | ⬝ | text-search-curie-doc-001 | 28. ⬝ text-search-curie-doc-001 | |||||||
34 | 29. | ⬝ | text-search-curie-query-001 | 29. ⬝ text-search-curie-query-001 | |||||||
35 | 30. | ▪ | text-search-davinci-doc-001 | 30. ▪ text-search-davinci-doc-001 | |||||||
36 | 31. | ▪ | text-search-davinci-query-001 | 31. ▪ text-search-davinci-query-001 | |||||||
37 | 32. | ⬝ | code-search-ada-code-001 | 32. ⬝ code-search-ada-code-001 | |||||||
38 | 33. | ⬝ | code-search-ada-text-001 | 33. ⬝ code-search-ada-text-001 | |||||||
39 | 34. | ⬝ | code-search-babbage-code-001 | 34. ⬝ code-search-babbage-code-001 | |||||||
40 | 35. | ⬝ | code-search-babbage-text-001 | 35. ⬝ code-search-babbage-text-001 | |||||||
41 | 36. | ▪ | code-cushman-001 | 12B | Codex | Aug/2021 | 36. ▪ code-cushman-001 12B Codex Aug/2021 | ||||
42 | 37. | ▪ | code-davinci-001 | 175B | Codex | Aug/2021 | 37. ▪ code-davinci-001 175B Codex Aug/2021 | ||||
43 | 38. | ▪ | 6.9B FIM | Jul/2022 | 38. ▪ 6.9B FIM Jul/2022 | ||||||
44 | ---- | InstructGPT | |||||||||
45 | 39. | ⬝ | GPT-3 1.3B pretrain | Mar/2022 | 39. ⬝ GPT-3 1.3B pretrain Mar/2022 | ||||||
46 | 40. | ⬝ | GPT-3 2.7B pretrain | Mar/2022 | 40. ⬝ GPT-3 2.7B pretrain Mar/2022 | ||||||
47 | 41. | ⬝ | GPT-3 6.7B pretrain | Mar/2022 | 41. ⬝ GPT-3 6.7B pretrain Mar/2022 | ||||||
48 | 42. | ⬝ | GPT-3 unsupervised cpt-text 1.2B | 42. ⬝ GPT-3 unsupervised cpt-text 1.2B | |||||||
49 | 43. | ⬝ | curie-instruct-beta 6.7B | 43. ⬝ curie-instruct-beta 6.7B | |||||||
50 | 44. | ▪ | davinci-instruct-beta | 175B | InstructGPT-3 (SFT) | 44. ▪ davinci-instruct-beta 175B InstructGPT-3 (SFT) | |||||
51 | 45. | ▪ | text-davinci-001 175B (FeedME) | 175B | (renamed) | 45. ▪ text-davinci-001 175B (FeedME) 175B (renamed) | |||||
52 | 46. | ⬝ | text-ada-001 | 350M | (FeedME) | 46. ⬝ text-ada-001 350M (FeedME) | |||||
53 | 47. | ⬝ | text-babbage-001 | 1.3B | (FeedME) | 47. ⬝ text-babbage-001 1.3B (FeedME) | |||||
54 | 48. | ⬝ | text-curie-001 | 6.7B | (FeedME) | 48. ⬝ text-curie-001 6.7B (FeedME) | |||||
55 | 49. | ▪ | text-davinci-001 | 175B | (FeedME) | 49. ▪ text-davinci-001 175B (FeedME) | |||||
56 | 50. | ▪ | text-davinci-insert-001 | 50. ▪ text-davinci-insert-001 | |||||||
57 | 51. | ▪ | text-davinci-edit-001 | 51. ▪ text-davinci-edit-001 | |||||||
58 | 52. | ▪ | code-davinci-edit-001 | 52. ▪ code-davinci-edit-001 | |||||||
59 | 53. | ⬝ | if-curie-v2 | 53. ⬝ if-curie-v2 | |||||||
60 | 54. | ▪ | if-davinci-v2 | 54. ▪ if-davinci-v2 | |||||||
61 | 55. | ▪ | if-davinci:3.0.0 | 55. ▪ if-davinci:3.0.0 | |||||||
62 | 56. | ▪ | davinci-if:3.0.0 | 56. ▪ davinci-if:3.0.0 | |||||||
63 | 57. | ▪ | davinci-instruct-beta:2.0.0 | (SFT) | 57. ▪ davinci-instruct-beta:2.0.0 (SFT) | ||||||
64 | 58. | ⬝ | text-moderation | (Jun/2022) | 58. ⬝ text-moderation (Jun/2022) | ||||||
65 | 59. | ⬝ | text-moderation-stable | (Jun/2022) | 59. ⬝ text-moderation-stable (Jun/2022) | ||||||
66 | 60. | ⬝ | text-ada:001 | 60. ⬝ text-ada:001 | |||||||
67 | 61. | ⬝ | text-babbage:001 | 61. ⬝ text-babbage:001 | |||||||
68 | 62. | ⬝ | text-curie:001 | 62. ⬝ text-curie:001 | |||||||
69 | 63. | ▪ | text-davinci:001 | 63. ▪ text-davinci:001 | |||||||
70 | 64. | ▪ | WebGPT | Dec/2021 | 64. ▪ WebGPT Dec/2021 | ||||||
71 | ---- | GPT-3.5 | |||||||||
72 | 65. | ▪ | text-davinci-002 | (FeedME) | 65. ▪ text-davinci-002 (FeedME) | ||||||
73 | 66. | ▪ | text-davinci-003 | (PPO) | Nov/2022 | 66. ▪ text-davinci-003 (PPO) Nov/2022 | |||||
74 | 67. | ▪ | code-davinci-002 | Codex (base) | 67. ▪ code-davinci-002 Codex (base) | ||||||
75 | 68. | ▪ | text-davinci-insert-002 | 68. ▪ text-davinci-insert-002 | |||||||
76 | 69. | ▪ | text-chat-davinci-002-20221122 | 69. ▪ text-chat-davinci-002-20221122 | |||||||
77 | 70. | ⬝ | gpt-3.5-turbo | 20B | (PPO) 💬 | [66.4] | Nov/2022 | 70. ⬝ gpt-3.5-turbo 20B (PPO) 💬 [66.4] Nov/2022 | |||
78 | 71. | ⬝ | gpt-3.5-turbo-instruct | 71. ⬝ gpt-3.5-turbo-instruct | |||||||
79 | 72. | ⬝ | gpt-3.5-turbo-0301 | 💬 | Mar/2023 | 72. ⬝ gpt-3.5-turbo-0301 💬 Mar/2023 | |||||
80 | 73. | ⬝ | gpt-3.5-turbo-0613 | 💬 | Jun/2023 | 73. ⬝ gpt-3.5-turbo-0613 💬 Jun/2023 | |||||
81 | 74. | ⬝ | gpt-3.5-turbo-16k | 74. ⬝ gpt-3.5-turbo-16k | 2025api | ||||||
82 | 75. | ⬝ | gpt-3.5-turbo-16k-0613 | Jun/2023 | 75. ⬝ gpt-3.5-turbo-16k-0613 Jun/2023 | ||||||
83 | 76. | ⬝ | gpt-3.5-turbo-1106 | 💬 | Nov/2023 | 76. ⬝ gpt-3.5-turbo-1106 💬 Nov/2023 | 2025api | ||||
84 | 77. | ⬝ | gpt-3.5-turbo-0125 | 💬 | Jan/2024 | 77. ⬝ gpt-3.5-turbo-0125 💬 Jan/2024 | 2025api | ||||
85 | 78. | ⬝ | Microsoft Sydney/Prometheus/GPT-3.5 | 78. ⬝ Microsoft Sydney/Prometheus/GPT-3.5 | |||||||
86 | 79. | ⬝ | babbage-002 | (Mar/2023) | 79. ⬝ babbage-002 (Mar/2023) | ||||||
87 | 80. | ▪ | davinci-002 | [70.2] | (Mar/2023) | 80. ▪ davinci-002 [70.2] (Mar/2023) | |||||
88 | ---- | GPT-4 | |||||||||
89 | 81. | ◾ | Microsoft Sydney/Prometheus/GPT-4 | (2022) | 81. ◾ Microsoft Sydney/Prometheus/GPT-4 (2022) | ||||||
90 | 82. | ◾ | GPT-4 Classic | 1760B | [86.4] | Mar/2023 | 82. ◾ GPT-4 Classic 1760B [86.4] Mar/2023 | ||||
91 | 83. | ◾ | gpt-4-0314 | Mar/2023 | 83. ◾ gpt-4-0314 Mar/2023 | ||||||
92 | 84. | ◾ | gpt-4-32k | Jun/2023 | 84. ◾ gpt-4-32k Jun/2023 | ||||||
93 | 85. | ◾ | gpt-4-0613 | Jun/2023 | 85. ◾ gpt-4-0613 Jun/2023 | ||||||
94 | 86. | ◾ | GPT-4V | (vision) | Nov/2023 | 86. ◾ GPT-4V (vision) Nov/2023 | |||||
95 | 87. | ◾ | gpt-4-1106-vision-preview | Nov/2023 | 87. ◾ gpt-4-1106-vision-preview Nov/2023 | ||||||
96 | 88. | ◾ | gpt-4-1106 | [84.7] | Nov/2023 | 88. ◾ gpt-4-1106 [84.7] Nov/2023 | 2025eval | ||||
97 | 89. | ◾ | gpt-4-0125 | [85.4] | Jan/2024 | 89. ◾ gpt-4-0125 [85.4] Jan/2024 | 2025eval | ||||
98 | 90. | ◾ | GPT-4 MathMix | May/2023 | 90. ◾ GPT-4 MathMix May/2023 | ||||||
99 | 91. | ⬝ | GPT-4b | 8B | Jan/2025 | 91. ⬝ GPT-4b 8B Jan/2025 | |||||
100 | 92. | ⬝ | GPT-4b micro | Jan/2025 | 92. ⬝ GPT-4b micro Jan/2025 |