Benchmark Almanac

General-Bench

Unified-io-2-XXL 13.7 tops its leaderboard InternVL2.5-8B 8.41 leads the models we track
Not in model cards

Models perform multimodal understanding and generation tasks. General-Level averages task scores and measures performance above specialist baselines at progressively broader levels.

Different tasks use different metrics. Higher level scores combine understanding, generation, and language scores rather than a single question-answer accuracy.

Released
May 2025
Kind
General visual questions
Level
University
Scored
Checked automatically
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Images, Video, Audio → Text, Images, Video, Audio
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Gemini 1.5 Pro Sep 2024 #157 on the capability index
Last model added
December 2024
Scores refreshed
November 20th, 2026

In the authors’ words

The Multimodal Large Language Model (MLLM) is currently experiencing rapid growth, driven by the advanced capabilities of LLMs. Unlike earlier specialists, existing MLLMs are evolving towards a Multimodal Generalist paradigm. Initially limited to understanding multiple modalities, these models have advanced to not only comprehend but also generate across modalities. Their capabilities have expanded from coarse-grained to fine-grained multimodal understanding and from supporting limited modalities to arbitrary ones. While many benchmarks exist to assess MLLMs, a critical question arises: Can we simply assume that higher performance across tasks indicates a stronger MLLM capability, bringing us closer to human-level AI? We argue that the answer is not as straightforward as it seems. This project introduces General-Level, an evaluation framework that defines 5-scale levels of MLLM performance and generality, offering a methodology to compare MLLMs and gauge the progress of existing systems towards more robust multimodal generalists and, ultimately, towards AGI. At the core of the framework is the concept of Synergy, which measures whether models maintain consistent capabilities across comprehension and generation, and across multiple modalities. To support this evaluation, we present General-Bench, which encompasses a broader spectrum of skills, modalities, formats, and capabilities, including over 700 tasks and 325,800 instances. The evaluation results that involve over 100 existing state-of-the-art MLLMs uncover the capability rankings of generalists, highlighting the challenges in reaching genuine AI. We expect this project to pave the way for future research on next-generation multimodal foundation models, providing a robust infrastructure to accelerate the realization of AGI. Project page: https://generalist.top/

Abstract of On Path to Multimodal Generalist: General-Level and General-Bench (arXiv:2505.04620, CC0)
101
Entries on its own leaderboard, 41 models we track

Unified-io-2-XXL tops the leaderboard with 13.7 InternVL2.5-8B leads with 8.41 among models released by December 2024

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is InternVL2.5-8B, at 8.41. This view shows only models we track, each at its best.

The newest model with a score here came out in December 2024, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

101 entries. Scroll sideways for the rest.

  1. 13.7 Unified-io-2-XXL
  2. 13.0 AnyGPT
  3. 13.0 NExT-GPT-V1.5
  4. 12.4 ImageBind-LLM
  5. 12.3 ModaVerse-7b-v0
  6. 12.1 Vitron-V1
  7. 11.8 PandaGPT-13B
  8. 10.8 VidAgent
  9. 8.41 InternVL2.5-8B
  10. 7.73 Emu2-37B
  11. 7.67 Sa2VA-26B
  12. 7.38 LaVIT-V2 (7B)
  13. 7.24 LLaVA-OneVision 72B
  14. 7.15 Qwen2-Audio-7B-Instruct
  15. 7.10 Qwen-Audio-Chat
  16. 6.97 Mini-Gemini
  17. 6.70 SEED-LLaMA-13B
  18. 6.59 GAMA
  19. 6.58 Qwen2-VL-72B
  20. 6.41 Sa2VA-8B
  21. 6.36 InternVL2.5-26B
  22. 6.11 Qwen2-VL-7B-Instruct
  23. 6.10 InternVL2.5-4B
  24. 5.99 SALMONN-13B
  25. 5.98 InternVL2-26B
  26. 5.87 Monkey-10B-chat
  27. 5.87 WavLLM
  28. 5.83 InternVL2.5-2B
  29. 5.82 Pengi
  30. 5.67 LLaVA-OneVision 7B
  31. 5.27 SALMONN-7B
  32. 4.97 DeepSeek-VL-7B-Chat
  33. 4.93 InternVL2-8B
  34. 4.92 Gemini 1.5 Pro
  35. 4.92 GPT-4o
  36. 4.92 GPT4-o-4096
  37. 4.85 Claude 3.5 Sonnet
  38. 4.75 Claude-3.5-Opus
  39. 4.74 chatgpt4-o-latest
  40. 4.64 Gemini 1.5 Flash
  41. 4.59 CoLVA-4B
  42. 4.54 GPT-4V
  43. 4.45 GPT-4o mini
  44. 4.25 GLM-VL-Chat
  45. 4.18 Idefics3-8B-Llama3
  46. 4.15 LLaVA-v1.6-34B
  47. 4.12 Phi-3.5-vision-instruct
  48. 4.12 MiniCPM3-4B
  49. 4.08 CogVLM-17B-Chat
  50. 4.05 CoLVA-2B
  51. 4.04 InternVL-Chat-V1-5
  52. 4.01 DetGPT
  53. 3.85 BLIP-3 (XGen-MM)
  54. 3.78 llava-v1.6-vicuna-13b
  55. 3.69 Pixtral 12B
  56. 3.68 ShareGPT4V-13B
  57. 3.65 Yi-vision-v2
  58. 3.48 Qwen-VL-Chat
  59. 3.45 ShareGPT4V-7B
  60. 3.38 Mini-InternVL-Chat-4B-V1-5
  61. 3.33 InternLM-XComposer2-VL-1.8B
  62. 3.28 DeepSeek-VL-7B-Base
  63. 3.22 MiniGPT4-LLaMA2-7B
  64. 3.14 MoE-LLAVA-Phi2-2.7B-4e-384
  65. 3.05 mPLUG-Owl2
  66. 2.94 Cambrian-1-8B
  67. 2.91 BLIP2
  68. 2.83 miniMonkey
  69. 2.66 NExT-Chat
  70. 2.20 Audio-GPT4
  71. 2.12 GPT4RoI-7B
  72. 1.95 Show-o
  73. 1.81 SpeechGPT-7B-com
  74. 1.75 PointLLM-13B
  75. 1.69 LM4LV
  76. 1.63 PointLLM-7B
  77. 1.46 Long-LLaVA-9B
  78. 1.35 3D-LLM-2.1B
  79. 1.35 3D-VisTA
  80. 1.14 OMG-LLaVA-InternLM20B
  81. 1.00 DeepSeek-VL2
  82. 0.91 DeepSeek-VL2-Small
  83. 0.79 Otter
  84. 0.40 LLaMA-mesh
  85. 0.32 LISA
  86. 0.24 GLaMM
  87. 0.05 AvatarGPT
  88. 0.00 ChatGLM-6B
  89. 0.00 GPT-3.5 Turbo
  90. 0.00 Gemma 2 9B
  91. 0.00 Llama 3.1-8B
  92. 0.00 Qwen2.5-7B
  93. 0.00 Yi-Lightning
  94. 0.00 Falcon3-7B-Instruct
  95. 0.00 Baichuan2-7B-Base
  96. 0.00 GPT-J
  97. 0.00 InternLM2-Chat-7B
  98. 0.00 Ministral-8B-Instruct-2410
  99. 0.00 MotionGPT-LLaMA
  100. 0.00 MotionGPT-T5
  101. 0.00 Vicuna-7b-V1.5
  1. 1 Unified-io-2-XXL 13.7
  2. 2 AnyGPT 13.0
  3. 3 NExT-GPT-V1.5 13.0
  4. 4 ImageBind-LLM 12.4
  5. 5 ModaVerse-7b-v0 12.3
  6. 6 Vitron-V1 12.1
  7. 7 PandaGPT-13B 11.8
  8. 8 VidAgent 10.8
  9. 9 InternVL2.5-8B 8.41
  10. 10 Emu2-37B 7.73
  11. 11 Sa2VA-26B 7.67
  12. 12 LaVIT-V2 (7B) 7.38
  13. 13 LLaVA-OneVision 72B 7.24
  14. 14 Qwen2-Audio-7B-Instruct 7.15
  15. 15 Qwen-Audio-Chat 7.10
  16. 16 Mini-Gemini 6.97
  17. 17 SEED-LLaMA-13B 6.70
  18. 18 GAMA 6.59
  19. 19 Qwen2-VL-72B 6.58
  20. 20 Sa2VA-8B 6.41
  21. 21 InternVL2.5-26B 6.36
  22. 22 Qwen2-VL-7B-Instruct 6.11
  23. 23 InternVL2.5-4B 6.10
  24. 24 SALMONN-13B 5.99
  25. 25 InternVL2-26B 5.98
  26. 26 Monkey-10B-chat 5.87
  27. 27 WavLLM 5.87
  28. 28 InternVL2.5-2B 5.83
  29. 29 Pengi 5.82
  30. 30 LLaVA-OneVision 7B 5.67
  31. 31 SALMONN-7B 5.27
  32. 32 DeepSeek-VL-7B-Chat 4.97
  33. 33 InternVL2-8B 4.93
  34. 34 Gemini 1.5 Pro Sep 2024 4.92
  35. 35 GPT-4o Nov 2024 4.92
  36. 36 GPT4-o-4096 4.92
  37. 37 Claude 3.5 Sonnet 4.85
  38. 38 Claude-3.5-Opus 4.75
  39. 39 chatgpt4-o-latest 4.74
  40. 40 Gemini 1.5 Flash Sep 2024 4.64
  41. 41 CoLVA-4B 4.59
  42. 42 GPT-4V 4.54
  43. 43 GPT-4o mini 4.45
  44. 44 GLM-VL-Chat 4.25
  45. 45 Idefics3-8B-Llama3 4.18
  46. 46 LLaVA-v1.6-34B 4.15
  47. 47 Phi-3.5-vision-instruct 4.12
  48. 48 MiniCPM3-4B 4.12
  49. 49 CogVLM-17B-Chat 4.08
  50. 50 CoLVA-2B 4.05
  51. 51 InternVL-Chat-V1-5 4.04
  52. 52 DetGPT 4.01
  53. 53 BLIP-3 (XGen-MM) 3.85
  54. 54 llava-v1.6-vicuna-13b 3.78
  55. 55 Pixtral 12B 3.69
  56. 56 ShareGPT4V-13B 3.68
  57. 57 Yi-vision-v2 3.65
  58. 58 Qwen-VL-Chat 3.48
  59. 59 ShareGPT4V-7B 3.45
  60. 60 Mini-InternVL-Chat-4B-V1-5 3.38
  61. 61 InternLM-XComposer2-VL-1.8B 3.33
  62. 62 DeepSeek-VL-7B-Base 3.28
  63. 63 MiniGPT4-LLaMA2-7B 3.22
  64. 64 MoE-LLAVA-Phi2-2.7B-4e-384 3.14
  65. 65 mPLUG-Owl2 3.05
  66. 66 Cambrian-1-8B 2.94
  67. 67 BLIP2 2.91
  68. 68 miniMonkey 2.83
  69. 69 NExT-Chat 2.66
  70. 70 Audio-GPT4 2.20
  71. 71 GPT4RoI-7B 2.12
  72. 72 Show-o 1.95
  73. 73 SpeechGPT-7B-com 1.81
  74. 74 PointLLM-13B 1.75
  75. 75 LM4LV 1.69
  76. 76 PointLLM-7B 1.63
  77. 77 Long-LLaVA-9B 1.46
  78. 78 3D-LLM-2.1B 1.35
  79. 79 3D-VisTA 1.35
  80. 80 OMG-LLaVA-InternLM20B 1.14
  81. 81 DeepSeek-VL2 1.00
  82. 82 DeepSeek-VL2-Small 0.91
  83. 83 Otter 0.79
  84. 84 LLaMA-mesh 0.40
  85. 85 LISA 0.32
  86. 86 GLaMM 0.24
  87. 87 AvatarGPT 0.05
  88. 88 ChatGLM-6B 0.00
  89. 89 GPT-3.5 Turbo Jan 2024 0.00
  90. 90 Gemma 2 9B 0.00
  91. 91 Llama 3.1-8B 0.00
  92. 92 Qwen2.5-7B 0.00
  93. 93 Yi-Lightning 0.00
  94. 94 Falcon3-7B-Instruct 0.00
  95. 95 Baichuan2-7B-Base 0.00
  96. 96 GPT-J 0.00
  97. 97 InternLM2-Chat-7B 0.00
  98. 98 Ministral-8B-Instruct-2410 0.00
  99. 99 MotionGPT-LLaMA 0.00
  100. 100 MotionGPT-T5 0.00
  101. 101 Vicuna-7b-V1.5 0.00

Source: Official General-Bench leaderboard

Source: Official General-Bench leaderboard

#EntryScoreSettingReleasedRun onSource
1 Unified-io-2-XXL 13.7 Dec 2024 Official leaderboard
2 AnyGPT 13.0 Dec 2024 Official leaderboard
3 NExT-GPT-V1.5 13.0 Dec 2024 Official leaderboard
4 ImageBind-LLM 12.4 Dec 2024 Official leaderboard
5 ModaVerse-7b-v0 12.3 Dec 2024 Official leaderboard
6 Vitron-V1 12.1 Dec 2024 Official leaderboard
7 PandaGPT-13B 11.8 Dec 2024 Official leaderboard
8 VidAgent 10.8 Dec 2024 Official leaderboard
9 InternVL2.5-8B 8.41 Dec 2024 Dec 2024 Official leaderboard
10 Emu2-37B 7.73 Dec 2024 Official leaderboard
11 Sa2VA-26B 7.67 Dec 2024 Official leaderboard
12 LaVIT-V2 (7B) 7.38 Dec 2024 Official leaderboard
13 LLaVA-OneVision 72B 7.24 Aug 2024 Dec 2024 Official leaderboard
14 Qwen2-Audio-7B-Instruct 7.15 Aug 2024 Dec 2024 Official leaderboard
15 Qwen-Audio-Chat 7.10 Dec 2024 Official leaderboard
16 Mini-Gemini 6.97 Dec 2024 Official leaderboard
17 SEED-LLaMA-13B 6.70 Dec 2024 Official leaderboard
18 GAMA 6.59 Dec 2024 Official leaderboard
19 Qwen2-VL-72B 6.58 Dec 2024 Official leaderboard
20 Sa2VA-8B 6.41 Dec 2024 Official leaderboard
21 InternVL2.5-26B 6.36 Dec 2024 Dec 2024 Official leaderboard
22 Qwen2-VL-7B-Instruct 6.11 Aug 2024 Dec 2024 Official leaderboard
23 InternVL2.5-4B 6.10 Dec 2024 Dec 2024 Official leaderboard
24 SALMONN-13B 5.99 Dec 2024 Official leaderboard
25 InternVL2-26B 5.98 Jul 2024 Dec 2024 Official leaderboard
26 Monkey-10B-chat 5.87 Dec 2024 Official leaderboard
27 WavLLM 5.87 Dec 2024 Official leaderboard
28 InternVL2.5-2B 5.83 Dec 2024 Dec 2024 Official leaderboard
29 Pengi 5.82 Dec 2024 Official leaderboard
30 LLaVA-OneVision 7B 5.67 Aug 2024 Dec 2024 Official leaderboard
31 SALMONN-7B 5.27 Dec 2024 Official leaderboard
32 DeepSeek-VL-7B-Chat 4.97 Mar 2024 Dec 2024 Official leaderboard
33 InternVL2-8B 4.93 Jul 2024 Dec 2024 Official leaderboard
34 Gemini 1.5 Pro Sep 2024 4.92 Sep 2024 Dec 2024 Official leaderboard
35 GPT-4o Nov 2024 4.92 Nov 2024 Dec 2024 Official leaderboard
36 GPT4-o-4096 4.92 Dec 2024 Official leaderboard
37 Claude 3.5 Sonnet 4.85 Jun 2024 Dec 2024 Official leaderboard
38 Claude-3.5-Opus 4.75 Dec 2024 Official leaderboard
39 chatgpt4-o-latest 4.74 Dec 2024 Official leaderboard
40 Gemini 1.5 Flash Sep 2024 4.64 Aug 2024 Dec 2024 Official leaderboard
41 CoLVA-4B 4.59 Dec 2024 Official leaderboard
42 GPT-4V 4.54 Nov 2023 Dec 2024 Official leaderboard
43 GPT-4o mini 4.45 Jul 2024 Dec 2024 Official leaderboard
44 GLM-VL-Chat 4.25 Dec 2024 Official leaderboard
45 Idefics3-8B-Llama3 4.18 Aug 2024 Dec 2024 Official leaderboard
46 LLaVA-v1.6-34B 4.15 Jan 2024 Dec 2024 Official leaderboard
47 Phi-3.5-vision-instruct 4.12 Aug 2024 Dec 2024 Official leaderboard
48 MiniCPM3-4B 4.12 Sep 2024 Dec 2024 Official leaderboard
49 CogVLM-17B-Chat 4.08 Nov 2023 Dec 2024 Official leaderboard
50 CoLVA-2B 4.05 Dec 2024 Official leaderboard
51 InternVL-Chat-V1-5 4.04 Apr 2024 Dec 2024 Official leaderboard
52 DetGPT 4.01 Dec 2024 Official leaderboard
53 BLIP-3 (XGen-MM) 3.85 Dec 2024 Official leaderboard
54 llava-v1.6-vicuna-13b 3.78 Jan 2024 Dec 2024 Official leaderboard
55 Pixtral 12B 3.69 Sep 2024 Dec 2024 Official leaderboard
56 ShareGPT4V-13B 3.68 Dec 2023 Dec 2024 Official leaderboard
57 Yi-vision-v2 3.65 Dec 2024 Official leaderboard
58 Qwen-VL-Chat 3.48 Aug 2023 Dec 2024 Official leaderboard
59 ShareGPT4V-7B 3.45 Nov 2023 Dec 2024 Official leaderboard
60 Mini-InternVL-Chat-4B-V1-5 3.38 May 2024 Dec 2024 Official leaderboard
61 InternLM-XComposer2-VL-1.8B 3.33 Apr 2024 Dec 2024 Official leaderboard
62 DeepSeek-VL-7B-Base 3.28 Dec 2024 Official leaderboard
63 MiniGPT4-LLaMA2-7B 3.22 Dec 2024 Official leaderboard
64 MoE-LLAVA-Phi2-2.7B-4e-384 3.14 Dec 2024 Official leaderboard
65 mPLUG-Owl2 3.05 Nov 2023 Dec 2024 Official leaderboard
66 Cambrian-1-8B 2.94 Jun 2024 Dec 2024 Official leaderboard
67 BLIP2 2.91 Dec 2024 Official leaderboard
68 miniMonkey 2.83 Dec 2024 Official leaderboard
69 NExT-Chat 2.66 Dec 2024 Official leaderboard
70 Audio-GPT4 2.20 Dec 2024 Official leaderboard
71 GPT4RoI-7B 2.12 Dec 2024 Official leaderboard
72 Show-o 1.95 Dec 2024 Official leaderboard
73 SpeechGPT-7B-com 1.81 Dec 2024 Official leaderboard
74 PointLLM-13B 1.75 Dec 2024 Official leaderboard
75 LM4LV 1.69 Dec 2024 Official leaderboard
76 PointLLM-7B 1.63 Dec 2024 Official leaderboard
77 Long-LLaVA-9B 1.46 Dec 2024 Official leaderboard
78 3D-LLM-2.1B 1.35 Dec 2024 Official leaderboard
79 3D-VisTA 1.35 Dec 2024 Official leaderboard
80 OMG-LLaVA-InternLM20B 1.14 Dec 2024 Official leaderboard
81 DeepSeek-VL2 1.00 Dec 2024 Dec 2024 Official leaderboard
82 DeepSeek-VL2-Small 0.91 Dec 2024 Dec 2024 Official leaderboard
83 Otter 0.79 Dec 2024 Official leaderboard
84 LLaMA-mesh 0.40 Dec 2024 Official leaderboard
85 LISA 0.32 Dec 2024 Official leaderboard
86 GLaMM 0.24 Dec 2024 Official leaderboard
87 AvatarGPT 0.05 Dec 2024 Official leaderboard
88 ChatGLM-6B 0.00 Mar 2023 Dec 2024 Official leaderboard
89 GPT-3.5 Turbo Jan 2024 0.00 Jan 2024 Dec 2024 Official leaderboard
90 Gemma 2 9B 0.00 Jun 2024 Dec 2024 Official leaderboard
91 Llama 3.1-8B 0.00 Jul 2024 Dec 2024 Official leaderboard
92 Qwen2.5-7B 0.00 Sep 2024 Dec 2024 Official leaderboard
93 Yi-Lightning 0.00 Dec 2024 Dec 2024 Official leaderboard
94 Falcon3-7B-Instruct 0.00 Dec 2024 Dec 2024 Official leaderboard
95 Baichuan2-7B-Base 0.00 Dec 2024 Official leaderboard
96 GPT-J 0.00 Dec 2024 Official leaderboard
97 InternLM2-Chat-7B 0.00 Dec 2024 Official leaderboard
98 Ministral-8B-Instruct-2410 0.00 Dec 2024 Official leaderboard
99 MotionGPT-LLaMA 0.00 Dec 2024 Official leaderboard
100 MotionGPT-T5 0.00 Dec 2024 Official leaderboard
101 Vicuna-7b-V1.5 0.00 Dec 2024 Official leaderboard
#ModelScoreSettingReleasedRun onSource
1 InternVL2.5-8B 8.41 Dec 2024 Dec 2024 Official leaderboard
2 LLaVA-OneVision 72B 7.24 Aug 2024 Dec 2024 Official leaderboard
3 Qwen2-Audio-7B-Instruct 7.15 Aug 2024 Dec 2024 Official leaderboard
4 InternVL2.5-26B 6.36 Dec 2024 Dec 2024 Official leaderboard
5 Qwen2-VL-7B-Instruct 6.11 Aug 2024 Dec 2024 Official leaderboard
6 InternVL2.5-4B 6.10 Dec 2024 Dec 2024 Official leaderboard
7 InternVL2-26B 5.98 Jul 2024 Dec 2024 Official leaderboard
8 InternVL2.5-2B 5.83 Dec 2024 Dec 2024 Official leaderboard
9 LLaVA-OneVision 7B 5.67 Aug 2024 Dec 2024 Official leaderboard
10 DeepSeek-VL-7B-Chat 4.97 Mar 2024 Dec 2024 Official leaderboard
11 InternVL2-8B 4.93 Jul 2024 Dec 2024 Official leaderboard
12 Gemini 1.5 Pro Sep 2024 4.92 Sep 2024 Dec 2024 Official leaderboard
13 GPT-4o Nov 2024 4.92 Nov 2024 Dec 2024 Official leaderboard
14 Claude 3.5 Sonnet 4.85 Jun 2024 Dec 2024 Official leaderboard
15 Gemini 1.5 Flash Sep 2024 4.64 Aug 2024 Dec 2024 Official leaderboard
16 GPT-4V 4.54 Nov 2023 Dec 2024 Official leaderboard
17 GPT-4o mini 4.45 Jul 2024 Dec 2024 Official leaderboard
18 Idefics3-8B-Llama3 4.18 Aug 2024 Dec 2024 Official leaderboard
19 LLaVA-v1.6-34B 4.15 Jan 2024 Dec 2024 Official leaderboard
20 Phi-3.5-vision-instruct 4.12 Aug 2024 Dec 2024 Official leaderboard
21 MiniCPM3-4B 4.12 Sep 2024 Dec 2024 Official leaderboard
22 CogVLM-17B-Chat 4.08 Nov 2023 Dec 2024 Official leaderboard
23 InternVL-Chat-V1-5 4.04 Apr 2024 Dec 2024 Official leaderboard
24 llava-v1.6-vicuna-13b 3.78 Jan 2024 Dec 2024 Official leaderboard
25 Pixtral 12B 3.69 Sep 2024 Dec 2024 Official leaderboard
26 ShareGPT4V-13B 3.68 Dec 2023 Dec 2024 Official leaderboard
27 Qwen-VL-Chat 3.48 Aug 2023 Dec 2024 Official leaderboard
28 ShareGPT4V-7B 3.45 Nov 2023 Dec 2024 Official leaderboard
29 Mini-InternVL-Chat-4B-V1-5 3.38 May 2024 Dec 2024 Official leaderboard
30 InternLM-XComposer2-VL-1.8B 3.33 Apr 2024 Dec 2024 Official leaderboard
31 mPLUG-Owl2 3.05 Nov 2023 Dec 2024 Official leaderboard
32 Cambrian-1-8B 2.94 Jun 2024 Dec 2024 Official leaderboard
33 DeepSeek-VL2 1.00 Dec 2024 Dec 2024 Official leaderboard
34 DeepSeek-VL2-Small 0.91 Dec 2024 Dec 2024 Official leaderboard
35 ChatGLM-6B 0.00 Mar 2023 Dec 2024 Official leaderboard
36 GPT-3.5 Turbo Jan 2024 0.00 Jan 2024 Dec 2024 Official leaderboard
37 Gemma 2 9B 0.00 Jun 2024 Dec 2024 Official leaderboard
38 Llama 3.1-8B 0.00 Jul 2024 Dec 2024 Official leaderboard
39 Qwen2.5-7B 0.00 Sep 2024 Dec 2024 Official leaderboard
40 Yi-Lightning 0.00 Dec 2024 Dec 2024 Official leaderboard
41 Falcon3-7B-Instruct 0.00 Dec 2024 Dec 2024 Official leaderboard

Rankings by part

We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Image, Video and Audio. InternVL2.5-8B leads 2 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

Image

41 models we track have a score here.

  1. 1 InternVL2.5-8B 25.2
  2. 2 InternVL2.5-4B ↑4 24.4
  3. 3 InternVL2.5-2B ↑5 23.3
  4. 4 LLaVA-OneVision 72B ↓2 23.1
  5. 5 DeepSeek-VL-7B-Chat ↑5 19.9
  6. 6 Gemini 1.5 Pro Sep 2024 ↑6 19.7
  7. 7 GPT-4o Nov 2024 ↑6 19.7
  8. 8 Claude 3.5 Sonnet ↑6 19.4
  9. 9 DeepSeek-VL2 ↑24 19.2
  10. 10 InternVL2.5-26B ↓6 18.7

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Shanghai AI Lab
Own leaderboard Not tested since May 2025

Source: Official General-Bench leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard
0.00 2.00 4.00 6.00 8.00 10.0 2023 2024 2025 0.00 2.00 4.00 6.00 8.00 10.0 2023 2024 2025

Source: Official General-Bench leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Vision

All 636 in Vision
  • MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. Qwen 3.6 Plus 86%
  • MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
  • CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
  • MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy. Seed2.1 Pro 90.7%
  • MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
  • MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy. Qwen 3.8 Max 85.9%

Data audit

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 23 months ago, in December 2024, for Falcon3-7B-Instruct. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (101 entries).
  • Pass Models scored 41 models we track, plus 60 other entries on its leaderboard.
  • Warning Newest score The newest came 23 months ago, in December 2024, for Falcon3-7B-Instruct.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 1.5 Pro (#157).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on generalist.top: 42 rows for models we track and 60 other entries at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check