OpenLMM Spatial Leaderboard
Models answer spatial reasoning tasks involving images. OpenCompass compares performance across the component evaluations.
The leaderboard aggregates results from component evaluations.
- Made by
- OpenCompass
- Released
- June 2025
- Kind
- Composite indices
- Level
- University
- Scored
- Settled by events
- Answers
- Short answer
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Items from
- Another benchmark
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision
- Strongest model tested
- GPT-5 #71 on the capability index
- Last model added
- August 2025
- Scores refreshed
- November 20th, 2026
GPT-5 leads with 62.5 among models released by August 2025 GPT-5 leads with 62.5 among models released by August 2025
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. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in August 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 62.5 GPT-5
- 57.3
Seed1.5-VL
- 57.1 GPT-5 mini
- 55.2 GPT-4.1
- 54.6 ChatGPT-4o-latest
- 54.3
Gemini 2.5 Flash
- 53.9 GPT-4o (1120, detail-high)
- 53.1 Qwen2-VL-72B-Instruct
- 52.8
InternVL3-38B
- 52.8
InternVL3-78B
- 52.0 Step-1o
- 51.8
Gemini 2.0 Flash
- 51.4
Qwen2.5-VL-72B
- 51.1 GPT-4.1 mini
- 51.0
Qwen2.5-VL-32B
- 50.8
Gemini 1.5 Pro
- 50.5 MiMo-VL-7B-RL
- 49.3 Seed1.6-Flash
- 49.2
InternVL2_5-78B
- 49.1 Claude 3.7 Sonnet
- 49.1
InternVL3-14B
- 49.0 Claude 3.5 Sonnet (October 2024)
- 48.9
Ovis2-34B
- 48.2 GPT-5 nano
- 47.6
Gemini 1.5 Flash
- 47.5
Ovis2-16B
- 47.4 GLM-4v-Plus-20250111
- 47.2 LLaVA-OneVision 72B
- 46.7
InternVL3-8B
- 45.9 Qwen2-VL-7B-Instruct
- 45.0
Gemini 2.5 Pro
- 44.4 Claude4-Sonnet
- 44.1
Gemma 3 27B
- 44.0 Kimi-VL-A3B-Instruct
- 43.7
SAIL-VL-1.6-8B
- 43.5
Ovis2-8B
- 43.3
InternVL2.5-8B
- 43.3
Qwen2.5-VL-7B
- 42.7
DeepSeek-VL2
- 42.3
Gemma 3 12B
- 42.1
Phi-4-multimodal
- 42.0 GPT-4o-mini (0718, detail-high)
- 41.7
InternLM-XComposer2.5-7B
- 41.5 MiniCPM-o-2_6
- 40.9
DeepSeek-VL2-Small
- 40.8
LLaVA-OneVision 7B
- 39.8 SAIL-VL-1.5-2B
- 39.5
Pixtral 12B
- 39.0 GPT-4.1 nano
- 38.5
Qwen2.5-VL-3B
- 38.4
InternVL3-2B
- 38.2 Phi-3.5-vision-instruct
- 38.1
InternVL3-1B
- 37.4
Gemma 3 4B
- 33.1
Idefics3-8B-Llama3
- 32.4
Janus-Pro-7B
- 30.1
SmolVLM2-2.2B-Instruct
- 19.9 llava-v1.6-vicuna-7b
- 19.6 llava-v1.6-vicuna-13b
- 1 GPT-5 Standard ✓ 62.5
- 2
Seed1.5-VL Standard ✓ 57.3
- 3 GPT-5 mini Standard ✓ 57.1
- 4 GPT-4.1 Standard ✓ 55.2
- 5 ChatGPT-4o-latest Standard ✓ 54.6
- 6
Gemini 2.5 Flash Jun 2025, Standard ✓ 54.3
- 7 GPT-4o (1120, detail-high) Standard ✓ 53.9
- 8 Qwen2-VL-72B-Instruct Standard ✓ 53.1
- 9
InternVL3-38B Standard ✓ 52.8
- 10
InternVL3-78B Standard ✓ 52.8
- 11 Step-1o Standard ✓ 52.0
- 12
Gemini 2.0 Flash Dec 2024, Standard ✓ 51.8
- 13
Qwen2.5-VL-72B Standard ✓ 51.4
- 14 GPT-4.1 mini Standard ✓ 51.1
- 15
Qwen2.5-VL-32B Standard ✓ 51.0
- 16
Gemini 1.5 Pro Sep 2024, Standard ✓ 50.8
- 17 MiMo-VL-7B-RL Standard ✓ 50.5
- 18 Seed1.6-Flash Standard ✓ 49.3
- 19
InternVL2_5-78B Standard ✓ 49.2
- 20 Claude 3.7 Sonnet Standard ✓ 49.1
- 21
InternVL3-14B Standard ✓ 49.1
- 22 Claude 3.5 Sonnet (October 2024) Standard ✓ 49.0
- 23
Ovis2-34B Standard ✓ 48.9
- 24 GPT-5 nano Standard ✓ 48.2
- 25
Gemini 1.5 Flash Sep 2024, Standard ✓ 47.6
- 26
Ovis2-16B Standard ✓ 47.5
- 27 GLM-4v-Plus-20250111 Standard ✓ 47.4
- 28 LLaVA-OneVision 72B Standard ✓ 47.2
- 29
InternVL3-8B Standard ✓ 46.7
- 30 Qwen2-VL-7B-Instruct Standard ✓ 45.9
- 31
Gemini 2.5 Pro Mar 2025, Circular ✓ 45.0
- 32 Claude4-Sonnet Standard ✓ 44.4
- 33
Gemma 3 27B Standard ✓ 44.1
- 34 Kimi-VL-A3B-Instruct Standard ✓ 44.0
- 35
SAIL-VL-1.6-8B Standard ✓ 43.7
- 36
Ovis2-8B Standard ✓ 43.5
- 37
InternVL2.5-8B Standard ✓ 43.3
- 38
Qwen2.5-VL-7B Standard ✓ 43.3
- 39
DeepSeek-VL2 Standard ✓ 42.7
- 40
Gemma 3 12B Standard ✓ 42.3
- 41
Phi-4-multimodal Standard ✓ 42.1
- 42 GPT-4o-mini (0718, detail-high) Standard ✓ 42.0
- 43
InternLM-XComposer2.5-7B Standard ✓ 41.7
- 44 MiniCPM-o-2_6 Standard ✓ 41.5
- 45
DeepSeek-VL2-Small Standard ✓ 40.9
- 46
LLaVA-OneVision 7B Standard ✓ 40.8
- 47 SAIL-VL-1.5-2B Standard ✓ 39.8
- 48
Pixtral 12B Standard ✓ 39.5
- 49 GPT-4.1 nano Standard ✓ 39.0
- 50
Qwen2.5-VL-3B Standard ✓ 38.5
- 51
InternVL3-2B Standard ✓ 38.4
- 52 Phi-3.5-vision-instruct Standard ✓ 38.2
- 53
InternVL3-1B Standard ✓ 38.1
- 54
Gemma 3 4B Standard ✓ 37.4
- 55
Idefics3-8B-Llama3 Standard ✓ 33.1
- 56
Janus-Pro-7B Standard ✓ 32.4
- 57
SmolVLM2-2.2B-Instruct Standard ✓ 30.1
- 58 llava-v1.6-vicuna-7b Standard ✓ 19.9
- 59 llava-v1.6-vicuna-13b Standard ✓ 19.6
Source: Official OpenLMM Spatial Leaderboard leaderboard
- 62.5 GPT-5
- 57.3
Seed1.5-VL
- 57.1 GPT-5 mini
- 55.2 GPT-4.1
- 54.3
Gemini 2.5 Flash
- 53.1 Qwen2-VL-72B-Instruct
- 52.8
InternVL3-38B
- 52.8
InternVL3-78B
- 51.8
Gemini 2.0 Flash
- 51.4
Qwen2.5-VL-72B
- 51.1 GPT-4.1 mini
- 51.0
Qwen2.5-VL-32B
- 50.8
Gemini 1.5 Pro
- 50.5 MiMo-VL-7B-RL
- 49.2
InternVL2_5-78B
- 49.1 Claude 3.7 Sonnet
- 49.1
InternVL3-14B
- 49.0 Claude 3.5 Sonnet (October 2024)
- 48.9
Ovis2-34B
- 48.2 GPT-5 nano
- 47.6
Gemini 1.5 Flash
- 47.5
Ovis2-16B
- 47.2 LLaVA-OneVision 72B
- 46.7
InternVL3-8B
- 45.9 Qwen2-VL-7B-Instruct
- 45.0
Gemini 2.5 Pro
- 44.1
Gemma 3 27B
- 44.0 Kimi-VL-A3B-Instruct
- 43.7
SAIL-VL-1.6-8B
- 43.5
Ovis2-8B
- 43.3
InternVL2.5-8B
- 43.3
Qwen2.5-VL-7B
- 42.7
DeepSeek-VL2
- 42.3
Gemma 3 12B
- 42.1
Phi-4-multimodal
- 41.7
InternLM-XComposer2.5-7B
- 41.5 MiniCPM-o-2_6
- 40.9
DeepSeek-VL2-Small
- 40.8
LLaVA-OneVision 7B
- 39.5
Pixtral 12B
- 39.0 GPT-4.1 nano
- 38.5
Qwen2.5-VL-3B
- 38.4
InternVL3-2B
- 38.2 Phi-3.5-vision-instruct
- 38.1
InternVL3-1B
- 37.4
Gemma 3 4B
- 33.1
Idefics3-8B-Llama3
- 32.4
Janus-Pro-7B
- 30.1
SmolVLM2-2.2B-Instruct
- 19.9 llava-v1.6-vicuna-7b
- 19.6 llava-v1.6-vicuna-13b
- 1 GPT-5 Standard ✓ 62.5
- 2
Seed1.5-VL Standard ✓ 57.3
- 3 GPT-5 mini Standard ✓ 57.1
- 4 GPT-4.1 Standard ✓ 55.2
- 5
Gemini 2.5 Flash Jun 2025, Standard ✓ 54.3
- 6 Qwen2-VL-72B-Instruct Standard ✓ 53.1
- 7
InternVL3-38B Standard ✓ 52.8
- 8
InternVL3-78B Standard ✓ 52.8
- 9
Gemini 2.0 Flash Dec 2024, Standard ✓ 51.8
- 10
Qwen2.5-VL-72B Standard ✓ 51.4
- 11 GPT-4.1 mini Standard ✓ 51.1
- 12
Qwen2.5-VL-32B Standard ✓ 51.0
- 13
Gemini 1.5 Pro Sep 2024, Standard ✓ 50.8
- 14 MiMo-VL-7B-RL Standard ✓ 50.5
- 15
InternVL2_5-78B Standard ✓ 49.2
- 16 Claude 3.7 Sonnet Standard ✓ 49.1
- 17
InternVL3-14B Standard ✓ 49.1
- 18 Claude 3.5 Sonnet (October 2024) Standard ✓ 49.0
- 19
Ovis2-34B Standard ✓ 48.9
- 20 GPT-5 nano Standard ✓ 48.2
- 21
Gemini 1.5 Flash Sep 2024, Standard ✓ 47.6
- 22
Ovis2-16B Standard ✓ 47.5
- 23 LLaVA-OneVision 72B Standard ✓ 47.2
- 24
InternVL3-8B Standard ✓ 46.7
- 25 Qwen2-VL-7B-Instruct Standard ✓ 45.9
- 26
Gemini 2.5 Pro Mar 2025, Circular ✓ 45.0
- 27
Gemma 3 27B Standard ✓ 44.1
- 28 Kimi-VL-A3B-Instruct Standard ✓ 44.0
- 29
SAIL-VL-1.6-8B Standard ✓ 43.7
- 30
Ovis2-8B Standard ✓ 43.5
- 31
InternVL2.5-8B Standard ✓ 43.3
- 32
Qwen2.5-VL-7B Standard ✓ 43.3
- 33
DeepSeek-VL2 Standard ✓ 42.7
- 34
Gemma 3 12B Standard ✓ 42.3
- 35
Phi-4-multimodal Standard ✓ 42.1
- 36
InternLM-XComposer2.5-7B Standard ✓ 41.7
- 37 MiniCPM-o-2_6 Standard ✓ 41.5
- 38
DeepSeek-VL2-Small Standard ✓ 40.9
- 39
LLaVA-OneVision 7B Standard ✓ 40.8
- 40
Pixtral 12B Standard ✓ 39.5
- 41 GPT-4.1 nano Standard ✓ 39.0
- 42
Qwen2.5-VL-3B Standard ✓ 38.5
- 43
InternVL3-2B Standard ✓ 38.4
- 44 Phi-3.5-vision-instruct Standard ✓ 38.2
- 45
InternVL3-1B Standard ✓ 38.1
- 46
Gemma 3 4B Standard ✓ 37.4
- 47
Idefics3-8B-Llama3 Standard ✓ 33.1
- 48
Janus-Pro-7B Standard ✓ 32.4
- 49
SmolVLM2-2.2B-Instruct Standard ✓ 30.1
- 50 llava-v1.6-vicuna-7b Standard ✓ 19.9
- 51 llava-v1.6-vicuna-13b Standard ✓ 19.6
Source: Official OpenLMM Spatial Leaderboard leaderboard
| # | Entry | Score | Setting | Released | Run on | Link | Source |
|---|---|---|---|---|---|---|---|
| 1 | GPT-5 ✓ | 62.5 | Standard | Aug 2025 | Aug 2025 | openai.com | Official leaderboard |
| 2 | | 57.3 | Standard | May 2025 | Jun 2025 | github.com | Official leaderboard |
| 3 | GPT-5 mini ✓ | 57.1 | Standard | Aug 2025 | Aug 2025 | openai.com | Official leaderboard |
| 4 | GPT-4.1 ✓ | 55.2 | Standard | Apr 2025 | Apr 2025 | openai.com | Official leaderboard |
| 5 | ChatGPT-4o-latest ✓ closed weights | 54.6 | Standard | Apr 2025 | openai.com | Official leaderboard | |
| 6 | | 54.3 | Standard | Jun 2025 | Jun 2025 | deepmind.google | Official leaderboard |
| 7 | GPT-4o (1120, detail-high) ✓ closed weights | 53.9 | Standard | Dec 2024 | openai.com | Official leaderboard | |
| 8 | Qwen2-VL-72B-Instruct ✓ | 53.1 | Standard | Aug 2024 | Oct 2024 | huggingface.co | Official leaderboard |
| 9 | | 52.8 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 10 | | 52.8 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 11 | Step-1o ✓ closed weights | 52.0 | Standard | Jan 2025 | platform.stepfun.com | Official leaderboard | |
| 12 | | 51.8 | Standard | Dec 2024 | Jan 2025 | deepmind.google | Official leaderboard |
| 13 | | 51.4 | Standard | Jan 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 14 | GPT-4.1 mini ✓ | 51.1 | Standard | Apr 2025 | Apr 2025 | openai.com | Official leaderboard |
| 15 | | 51.0 | Standard | Mar 2025 | Jun 2025 | huggingface.co | Official leaderboard |
| 16 | | 50.8 | Standard | Sep 2024 | Dec 2024 | deepmind.google | Official leaderboard |
| 17 | MiMo-VL-7B-RL ✓ | 50.5 | Standard | May 2025 | Jun 2025 | huggingface.co | Official leaderboard |
| 18 | Seed1.6-Flash ✓ closed weights | 49.3 | Standard | Jun 2025 | seed.bytedance.com | Official leaderboard | |
| 19 | | 49.2 | Standard | Dec 2024 | Dec 2024 | huggingface.co | Official leaderboard |
| 20 | Claude 3.7 Sonnet ✓ | 49.1 | Standard | Feb 2025 | Feb 2025 | docs.anthropic.com | Official leaderboard |
| 21 | | 49.1 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 22 | Claude 3.5 Sonnet (October 2024) ✓ | 49.0 | Standard | Oct 2024 | Nov 2024 | docs.anthropic.com | Official leaderboard |
| 23 | | 48.9 | Standard | Feb 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 24 | GPT-5 nano ✓ | 48.2 | Standard | Aug 2025 | Aug 2025 | openai.com | Official leaderboard |
| 25 | | 47.6 | Standard | Aug 2024 | Dec 2024 | deepmind.google | Official leaderboard |
| 26 | | 47.5 | Standard | Feb 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 27 | GLM-4v-Plus-20250111 ✓ closed weights | 47.4 | Standard | Jan 2025 | open.bigmodel.cn | Official leaderboard | |
| 28 | LLaVA-OneVision 72B ✓ | 47.2 | Standard | Aug 2024 | Sep 2024 | huggingface.co | Official leaderboard |
| 29 | | 46.7 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 30 | Qwen2-VL-7B-Instruct ✓ | 45.9 | Standard | Aug 2024 | Sep 2024 | huggingface.co | Official leaderboard |
| 31 | | 45.0 | Circular | Mar 2025 | Apr 2025 | deepmind.google | Official leaderboard |
| 32 | Claude4-Sonnet ✓ closed weights | 44.4 | Standard | Jun 2025 | anthropic.com | Official leaderboard | |
| 33 | | 44.1 | Standard | Mar 2025 | Mar 2025 | huggingface.co | Official leaderboard |
| 34 | Kimi-VL-A3B-Instruct ✓ | 44.0 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 35 | | 43.7 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 36 | | 43.5 | Standard | Feb 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 37 | | 43.3 | Standard | Dec 2024 | Dec 2024 | huggingface.co | Official leaderboard |
| 38 | | 43.3 | Standard | Jan 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 39 | | 42.7 | Standard | Dec 2024 | Jan 2025 | huggingface.co | Official leaderboard |
| 40 | | 42.3 | Standard | Mar 2025 | Mar 2025 | huggingface.co | Official leaderboard |
| 41 | | 42.1 | Standard | Feb 2025 | Mar 2025 | huggingface.co | Official leaderboard |
| 42 | GPT-4o-mini (0718, detail-high) ✓ closed weights | 42.0 | Standard | Jul 2024 | openai.com | Official leaderboard | |
| 43 | | 41.7 | Standard | Jul 2024 | Jul 2024 | huggingface.co | Official leaderboard |
| 44 | MiniCPM-o-2_6 ✓ | 41.5 | Standard | Jan 2025 | Jan 2025 | huggingface.co | Official leaderboard |
| 45 | | 40.9 | Standard | Dec 2024 | Jan 2025 | huggingface.co | Official leaderboard |
| 46 | | 40.8 | Standard | Aug 2024 | Sep 2024 | huggingface.co | Official leaderboard |
| 47 | SAIL-VL-1.5-2B ✓ open weights, 2.5B | 39.8 | Standard | Apr 2025 | huggingface.co | Official leaderboard | |
| 48 | | 39.5 | Standard | Sep 2024 | Sep 2024 | huggingface.co | Official leaderboard |
| 49 | GPT-4.1 nano ✓ | 39.0 | Standard | Apr 2025 | Apr 2025 | openai.com | Official leaderboard |
| 50 | | 38.5 | Standard | Jan 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 51 | | 38.4 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 52 | Phi-3.5-vision-instruct ✓ | 38.2 | Standard | Aug 2024 | Aug 2024 | huggingface.co | Official leaderboard |
| 53 | | 38.1 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 54 | | 37.4 | Standard | Mar 2025 | Mar 2025 | huggingface.co | Official leaderboard |
| 55 | | 33.1 | Standard | Aug 2024 | Aug 2024 | huggingface.co | Official leaderboard |
| 56 | | 32.4 | Standard | Jan 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 57 | | 30.1 | Standard | Feb 2025 | Mar 2025 | huggingface.co | Official leaderboard |
| 58 | llava-v1.6-vicuna-7b ✓ | 19.9 | Standard | Jan 2024 | Mar 2024 | llava-vl.github.io | Official leaderboard |
| 59 | llava-v1.6-vicuna-13b ✓ | 19.6 | Standard | Jan 2024 | Mar 2024 | llava-vl.github.io | Official leaderboard |
| # | Model | Score | Setting | Released | Run on | Link | Source |
|---|---|---|---|---|---|---|---|
| 1 | GPT-5 ✓ | 62.5 | Standard | Aug 2025 | Aug 2025 | openai.com | Official leaderboard |
| 2 | | 57.3 | Standard | May 2025 | Jun 2025 | github.com | Official leaderboard |
| 3 | GPT-5 mini ✓ | 57.1 | Standard | Aug 2025 | Aug 2025 | openai.com | Official leaderboard |
| 4 | GPT-4.1 ✓ | 55.2 | Standard | Apr 2025 | Apr 2025 | openai.com | Official leaderboard |
| 5 | | 54.3 | Standard | Jun 2025 | Jun 2025 | deepmind.google | Official leaderboard |
| 6 | Qwen2-VL-72B-Instruct ✓ | 53.1 | Standard | Aug 2024 | Oct 2024 | huggingface.co | Official leaderboard |
| 7 | | 52.8 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 8 | | 52.8 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 9 | | 51.8 | Standard | Dec 2024 | Jan 2025 | deepmind.google | Official leaderboard |
| 10 | | 51.4 | Standard | Jan 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 11 | GPT-4.1 mini ✓ | 51.1 | Standard | Apr 2025 | Apr 2025 | openai.com | Official leaderboard |
| 12 | | 51.0 | Standard | Mar 2025 | Jun 2025 | huggingface.co | Official leaderboard |
| 13 | | 50.8 | Standard | Sep 2024 | Dec 2024 | deepmind.google | Official leaderboard |
| 14 | MiMo-VL-7B-RL ✓ | 50.5 | Standard | May 2025 | Jun 2025 | huggingface.co | Official leaderboard |
| 15 | | 49.2 | Standard | Dec 2024 | Dec 2024 | huggingface.co | Official leaderboard |
| 16 | Claude 3.7 Sonnet ✓ | 49.1 | Standard | Feb 2025 | Feb 2025 | docs.anthropic.com | Official leaderboard |
| 17 | | 49.1 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 18 | Claude 3.5 Sonnet (October 2024) ✓ | 49.0 | Standard | Oct 2024 | Nov 2024 | docs.anthropic.com | Official leaderboard |
| 19 | | 48.9 | Standard | Feb 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 20 | GPT-5 nano ✓ | 48.2 | Standard | Aug 2025 | Aug 2025 | openai.com | Official leaderboard |
| 21 | | 47.6 | Standard | Aug 2024 | Dec 2024 | deepmind.google | Official leaderboard |
| 22 | | 47.5 | Standard | Feb 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 23 | LLaVA-OneVision 72B ✓ | 47.2 | Standard | Aug 2024 | Sep 2024 | huggingface.co | Official leaderboard |
| 24 | | 46.7 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 25 | Qwen2-VL-7B-Instruct ✓ | 45.9 | Standard | Aug 2024 | Sep 2024 | huggingface.co | Official leaderboard |
| 26 | | 45.0 | Circular | Mar 2025 | Apr 2025 | deepmind.google | Official leaderboard |
| 27 | | 44.1 | Standard | Mar 2025 | Mar 2025 | huggingface.co | Official leaderboard |
| 28 | Kimi-VL-A3B-Instruct ✓ | 44.0 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 29 | | 43.7 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 30 | | 43.5 | Standard | Feb 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 31 | | 43.3 | Standard | Dec 2024 | Dec 2024 | huggingface.co | Official leaderboard |
| 32 | | 43.3 | Standard | Jan 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 33 | | 42.7 | Standard | Dec 2024 | Jan 2025 | huggingface.co | Official leaderboard |
| 34 | | 42.3 | Standard | Mar 2025 | Mar 2025 | huggingface.co | Official leaderboard |
| 35 | | 42.1 | Standard | Feb 2025 | Mar 2025 | huggingface.co | Official leaderboard |
| 36 | | 41.7 | Standard | Jul 2024 | Jul 2024 | huggingface.co | Official leaderboard |
| 37 | MiniCPM-o-2_6 ✓ | 41.5 | Standard | Jan 2025 | Jan 2025 | huggingface.co | Official leaderboard |
| 38 | | 40.9 | Standard | Dec 2024 | Jan 2025 | huggingface.co | Official leaderboard |
| 39 | | 40.8 | Standard | Aug 2024 | Sep 2024 | huggingface.co | Official leaderboard |
| 40 | | 39.5 | Standard | Sep 2024 | Sep 2024 | huggingface.co | Official leaderboard |
| 41 | GPT-4.1 nano ✓ | 39.0 | Standard | Apr 2025 | Apr 2025 | openai.com | Official leaderboard |
| 42 | | 38.5 | Standard | Jan 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 43 | | 38.4 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 44 | Phi-3.5-vision-instruct ✓ | 38.2 | Standard | Aug 2024 | Aug 2024 | huggingface.co | Official leaderboard |
| 45 | | 38.1 | Standard | Apr 2025 | Apr 2025 | huggingface.co | Official leaderboard |
| 46 | | 37.4 | Standard | Mar 2025 | Mar 2025 | huggingface.co | Official leaderboard |
| 47 | | 33.1 | Standard | Aug 2024 | Aug 2024 | huggingface.co | Official leaderboard |
| 48 | | 32.4 | Standard | Jan 2025 | Feb 2025 | huggingface.co | Official leaderboard |
| 49 | | 30.1 | Standard | Feb 2025 | Mar 2025 | huggingface.co | Official leaderboard |
| 50 | llava-v1.6-vicuna-7b ✓ | 19.9 | Standard | Jan 2024 | Mar 2024 | llava-vl.github.io | Official leaderboard |
| 51 | llava-v1.6-vicuna-13b ✓ | 19.6 | Standard | Jan 2024 | Mar 2024 | llava-vl.github.io | Official leaderboard |
Rankings by part
We read 60 parts from the benchmark’s own leaderboard in 5 groups: blink, lego, 3dsrbench, spatial457 and mmsibench. Each ranks models on one part of the test. GPT-5 leads 23 of them, and 15 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
BLINK
51 models we track have a score here.
- 1
Seed1.5-VL ↑1 72.8
- 2 GPT-5 ↓1 70.1
- 3 GPT-4.1 ↑1 68.0
- 4 GPT-5 mini ↓1 67.0
- 5
Gemini 2.5 Flash Jun 2025 66.6
- 6
InternVL3-78B ↑2 66.5
- 7 GPT-4.1 mini ↑4 64.6
- 8
InternVL3-38B ↓1 64.5
- 9
Gemini 2.0 Flash Dec 2024 63.9
- 10
Gemini 1.5 Pro Sep 2024 ↑3 63.8
LEGO
51 models we track have a score here.
- 1 GPT-5 71.6
- 2
Seed1.5-VL 62.7
- 3 GPT-5 mini 61.5
- 4 GPT-4.1 59.3
- 5
Gemini 2.5 Flash Jun 2025 57.3
- 6 Claude 3.7 Sonnet ↑10 55.2
- 7 Claude 3.5 Sonnet (October 2024) ↑11 54.5
- 8
Gemini 2.0 Flash Dec 2024 ↑1 53.9
- 9
Qwen2.5-VL-72B ↑1 52.4
- 10
Gemini 1.5 Pro Sep 2024 ↑3 52.1
3DSRBench
51 models we track have a score here.
- 1 GPT-5 66.7
- 2
Seed1.5-VL 64.0
- 3 GPT-5 mini 64.0
- 4
Gemini 2.5 Flash Jun 2025 ↑1 61.5
- 5 GPT-5 nano ↑15 61.5
- 6
InternVL3-78B ↑2 61.3
- 7 GPT-4.1 mini ↑4 60.6
- 8 MiMo-VL-7B-RL ↑6 60.1
- 9 GPT-4.1 ↓5 59.2
- 10
InternVL3-38B ↓3 59.1
Spatial457
50 models we track have a score here.
- 1 Qwen2-VL-72B-Instruct ↑5 67.4
- 2
InternVL3-38B ↑5 64.3
- 3 Qwen2-VL-7B-Instruct ↑22 62.6
- 4 GPT-5 ↓3 62.3
- 5
InternVL3-14B ↑12 60.9
- 6
InternVL3-78B ↑2 60.4
- 7
Ovis2-34B ↑12 59.6
- 8 GPT-5 mini ↓5 58.0
- 9 GPT-4.1 ↓5 57.2
- 10
InternVL3-8B ↑14 56.6
MMSIBench
51 models we track have a score here.
- 1 GPT-5 41.9
- 2 GPT-5 mini ↑1 34.9
- 3
Gemini 2.5 Flash Jun 2025 ↑2 33.1
- 4 GPT-4.1 32.5
- 5 Qwen2-VL-72B-Instruct ↑1 32.2
- 6
InternVL2_5-78B ↑9 31.5
- 7
Qwen2.5-VL-72B ↑3 31.5
- 8
Ovis2-16B ↑14 31.4
- 9 Claude 3.5 Sonnet (October 2024) ↑9 31.3
- 10 GPT-4.1 nano ↑31 31.3
Art_Style
51 models we track have a score here.
- 1 Phi-3.5-vision-instruct ↑43 87.2
- 2
Phi-4-multimodal ↑33 87.2
- 3
Gemini 2.5 Pro Mar 2025 ↑23 84.6
- 4
Gemini 2.5 Flash Jun 2025 ↑1 84.6
- 5 GPT-5 ↓4 84.6
- 6 GPT-5 mini ↓3 83.8
- 7 GPT-4.1 ↓3 82.9
- 8
Seed1.5-VL ↓6 82.9
- 9
Gemini 1.5 Pro Sep 2024 ↑4 80.3
- 10 Claude 3.5 Sonnet (October 2024) ↑8 80.3
Counting
51 models we track have a score here.
- 1
Seed1.5-VL ↑1 81.7
- 2 Qwen2-VL-72B-Instruct ↑4 80.0
- 3
SAIL-VL-1.6-8B ↑26 80.0
- 4
InternVL3-78B ↑4 78.3
- 5
InternVL3-38B ↑2 77.5
- 6 MiMo-VL-7B-RL ↑8 77.5
- 7
Ovis2-8B ↑23 76.7
- 8 GPT-5 ↓7 76.7
- 9 Qwen2-VL-7B-Instruct ↑16 75.8
- 10
Ovis2-16B ↑12 75.8
Forensic_Detection
51 models we track have a score here.
- 1 GPT-5 90.9
- 2 Phi-3.5-vision-instruct ↑42 90.2
- 3
Phi-4-multimodal ↑32 90.2
- 4 GPT-4.1 89.4
- 5
Seed1.5-VL ↓3 89.4
- 6
Gemini 2.5 Pro Mar 2025 ↑20 87.1
- 7
Gemini 2.5 Flash Jun 2025 ↓2 82.6
- 8 GPT-5 mini ↓5 80.3
- 9 GPT-4.1 mini ↑2 78.8
- 10
Qwen2.5-VL-32B ↑2 76.5
Functional_Correspondence
51 models we track have a score here.
- 1 GPT-5 63.8
- 2 GPT-5 mini ↑1 61.5
- 3
Seed1.5-VL ↓1 60.0
- 4
Gemini 2.5 Flash Jun 2025 ↑1 59.2
- 5
Gemini 2.0 Flash Dec 2024 ↑4 58.5
- 6 GPT-4.1 ↓2 51.5
- 7
Gemini 1.5 Pro Sep 2024 ↑6 46.9
- 8
InternVL3-78B 46.9
- 9
Gemini 1.5 Flash Sep 2024 ↑12 42.3
- 10
Gemini 2.5 Pro Mar 2025 ↑16 42.3
IQ_Test
51 models we track have a score here.
- 1
InternVL2.5-8B ↑30 35.3
- 2
Qwen2.5-VL-32B ↑10 34.0
- 3
Seed1.5-VL ↓1 34.0
- 4
Ovis2-16B ↑18 33.3
- 5
Gemini 1.5 Pro Sep 2024 ↑8 32.7
- 6 GPT-5 ↓5 32.7
- 7
InternVL3-78B ↑1 32.0
- 8 GPT-5 mini ↓5 32.0
- 9
InternLM-XComposer2.5-7B ↑27 30.0
- 10
LLaVA-OneVision 7B ↑29 30.0
Jigsaw
51 models we track have a score here.
- 1
InternVL3-78B ↑7 85.3
- 2
SAIL-VL-1.6-8B ↑27 84.7
- 3
Qwen2.5-VL-32B ↑9 82.7
- 4 Kimi-VL-A3B-Instruct ↑24 82.7
- 5
InternVL2_5-78B ↑10 81.3
- 6
InternVL3-8B ↑18 81.3
- 7
InternVL3-38B 76.7
- 8
Seed1.5-VL ↓6 75.3
- 9
Ovis2-8B ↑21 74.7
- 10 LLaVA-OneVision 72B ↑13 74.0
Multi-view_Reasoning
51 models we track have a score here.
- 1
Phi-4-multimodal ↑34 80.5
- 2 Kimi-VL-A3B-Instruct ↑26 57.9
- 3
InternVL3-2B ↑40 56.4
- 4
InternLM-XComposer2.5-7B ↑32 55.6
- 5 LLaVA-OneVision 72B ↑18 55.6
- 6
LLaVA-OneVision 7B ↑33 55.6
- 7 Qwen2-VL-7B-Instruct ↑18 55.6
- 8
Gemini 1.5 Pro Sep 2024 ↑5 55.6
- 9
InternVL2_5-78B ↑6 55.6
- 10 MiniCPM-o-2_6 ↑27 55.6
Object_Localization
51 models we track have a score here.
- 1
Seed1.5-VL ↑1 77.9
- 2
InternVL3-38B ↑5 73.8
- 3 GPT-4.1 ↑1 72.1
- 4
Gemini 1.5 Pro Sep 2024 ↑9 71.3
- 5
InternVL3-14B ↑12 69.7
- 6 GPT-5 ↓5 69.7
- 7
Gemini 1.5 Flash Sep 2024 ↑14 67.2
- 8
DeepSeek-VL2 ↑25 67.2
- 9 Kimi-VL-A3B-Instruct ↑19 67.2
- 10 LLaVA-OneVision 72B ↑13 65.6
Relative_Depth
51 models we track have a score here.
- 1
InternVL3-14B ↑16 89.5
- 2
Seed1.5-VL 88.7
- 3
Qwen2.5-VL-72B ↑7 86.3
- 4 GPT-5 ↓3 86.3
- 5
InternVL2_5-78B ↑10 83.9
- 6 GPT-5 mini ↓3 83.9
- 7 Qwen2-VL-72B-Instruct ↓1 83.1
- 8
Gemini 2.5 Pro Mar 2025 ↑18 83.1
- 9
InternVL3-78B ↓1 83.1
- 10 MiMo-VL-7B-RL ↑4 83.1
Relative_Reflectance
51 models we track have a score here.
- 1
InternVL2.5-8B ↑30 39.6
- 2
Qwen2.5-VL-3B ↑40 39.6
- 3
Qwen2.5-VL-7B ↑29 39.6
- 4
InternVL3-38B ↑3 39.6
- 5 llava-v1.6-vicuna-13b ↑46 38.8
- 6 Qwen2-VL-72B-Instruct 38.8
- 7 Qwen2-VL-7B-Instruct ↑18 38.8
- 8
SmolVLM2-2.2B-Instruct ↑41 38.8
- 9
Ovis2-16B ↑13 38.8
- 10
Idefics3-8B-Llama3 ↑37 38.1
Semantic_Correspondence
51 models we track have a score here.
- 1
Seed1.5-VL ↑1 74.8
- 2 GPT-4.1 mini ↑9 70.5
- 3 GPT-5 ↓2 68.3
- 4 GPT-5 mini ↓1 64.7
- 5 GPT-4.1 ↓1 63.3
- 6
Gemini 2.5 Flash Jun 2025 ↓1 60.4
- 7
Gemini 2.0 Flash Dec 2024 ↑2 59.0
- 8
Gemini 1.5 Pro Sep 2024 ↑5 54.0
- 9 GPT-5 nano ↑11 53.2
- 10
InternVL3-38B ↓3 52.5
Spatial_Relation
51 models we track have a score here.
- 1
InternVL3-78B ↑7 93.7
- 2
InternVL3-14B ↑15 91.6
- 3
InternVL3-8B ↑21 90.9
- 4 MiniCPM-o-2_6 ↑33 89.5
- 5
Seed1.5-VL ↓3 89.5
- 6
Qwen2.5-VL-7B ↑26 88.8
- 7
InternVL3-38B 88.8
- 8
Gemini 2.5 Flash Jun 2025 ↓3 88.8
- 9 GPT-5 ↓8 88.8
- 10
InternVL2_5-78B ↑5 88.1
Visual_Correspondence
51 models we track have a score here.
- 1 GPT-4.1 ↑3 94.8
- 2
Seed1.5-VL 94.8
- 3 GPT-5 ↓2 93.6
- 4 GPT-4.1 mini ↑7 90.1
- 5 GPT-5 mini ↓2 89.5
- 6
InternVL3-78B ↑2 89.0
- 7
Gemini 2.0 Flash Dec 2024 ↑2 87.8
- 8
Gemini 2.5 Flash Jun 2025 ↓3 87.2
- 9
Gemini 2.5 Pro Mar 2025 ↑17 84.3
- 10
InternVL3-38B ↓3 84.3
Visual_Similarity
51 models we track have a score here.
- 1
InternVL2_5-78B ↑14 91.9
- 2 MiniCPM-o-2_6 ↑35 91.9
- 3
InternVL3-78B ↑5 91.9
- 4 Qwen2-VL-72B-Instruct ↑2 91.1
- 5
InternVL2.5-8B ↑26 91.1
- 6
Ovis2-34B ↑13 91.1
- 7
InternVL3-38B 91.1
- 8 Claude 3.7 Sonnet ↑8 90.4
- 9
Seed1.5-VL ↓7 90.4
- 10 Kimi-VL-A3B-Instruct ↑18 89.6
adjacency
51 models we track have a score here.
- 1 GPT-5 68.0
- 2
Pixtral 12B ↑38 67.0
- 3 GPT-4.1 ↑1 66.0
- 4 Qwen2-VL-72B-Instruct ↑2 64.0
- 5
Gemini 2.0 Flash Dec 2024 ↑4 64.0
- 6
InternVL2_5-78B ↑9 64.0
- 7
Gemma 3 12B ↑27 64.0
- 8 MiMo-VL-7B-RL ↑6 64.0
- 9
Gemini 1.5 Pro Sep 2024 ↑4 63.0
- 10
Gemini 2.5 Flash Jun 2025 ↓5 63.0
backwards
51 models we track have a score here.
- 1
Seed1.5-VL ↑1 88.0
- 2 MiMo-VL-7B-RL ↑12 78.0
- 3 GPT-5 mini 76.0
- 4
Qwen2.5-VL-72B ↑6 74.0
- 5 GPT-5 ↓4 72.0
- 6
Qwen2.5-VL-32B ↑6 66.0
- 7 Claude 3.7 Sonnet ↑9 63.0
- 8 GPT-4.1 mini ↑3 63.0
- 9
Gemini 2.0 Flash Dec 2024 58.0
- 10 GPT-4.1 ↓6 58.0
dependency
51 models we track have a score here.
- 1 GPT-5 94.0
- 2 GPT-5 mini ↑1 89.0
- 3 GPT-4.1 ↑1 88.0
- 4
Gemini 2.5 Flash Jun 2025 ↑1 85.0
- 5 GPT-4.1 mini ↑6 84.0
- 6
Gemini 1.5 Pro Sep 2024 ↑7 83.0
- 7 Claude 3.7 Sonnet ↑9 83.0
- 8
InternVL3-38B ↓1 81.0
- 9
Gemini 2.0 Flash Dec 2024 80.0
- 10
InternVL2_5-78B ↑5 80.0
height
51 models we track have a score here.
- 1 GPT-5 65.0
- 2
Seed1.5-VL 54.0
- 3 GPT-4.1 ↑1 51.0
- 4
Gemini 2.5 Flash Jun 2025 ↑1 50.0
- 5
InternVL3-78B ↑3 48.0
- 6 LLaVA-OneVision 72B ↑17 47.0
- 7
Gemini 2.5 Pro Mar 2025 ↑19 46.0
- 8 GPT-5 nano ↑12 45.0
- 9
LLaVA-OneVision 7B ↑30 44.0
- 10 GPT-4.1 mini ↑1 43.0
multi_view
51 models we track have a score here.
- 1 GPT-5 69.0
- 2 GPT-5 mini ↑1 69.0
- 3
Gemini 2.5 Flash Jun 2025 ↑2 60.0
- 4
Seed1.5-VL ↓2 57.0
- 5
InternVL3-38B ↑2 55.0
- 6
InternVL3-78B ↑2 54.0
- 7 Claude 3.5 Sonnet (October 2024) ↑11 53.0
- 8
Gemini 1.5 Pro Sep 2024 ↑5 52.0
- 9 Claude 3.7 Sonnet ↑7 52.0
- 10
Qwen2.5-VL-32B ↑2 52.0
next_step
51 models we track have a score here.
- 1 GPT-5 80.0
- 2
Gemini 2.0 Flash Dec 2024 ↑7 67.0
- 3 GPT-4.1 ↑1 65.0
- 4
Gemini 2.5 Flash Jun 2025 ↑1 65.0
- 5 GPT-5 mini ↓2 65.0
- 6
InternVL3-78B ↑2 64.0
- 7 Claude 3.5 Sonnet (October 2024) ↑11 63.0
- 8
InternVL2_5-78B ↑7 63.0
- 9
Gemini 2.5 Pro Mar 2025 ↑17 63.0
- 10 Claude 3.7 Sonnet ↑6 62.0
ordering
51 models we track have a score here.
- 1 GPT-5 86.0
- 2
Gemini 2.5 Pro Mar 2025 ↑24 74.0
- 3
Seed1.5-VL ↓1 69.0
- 4 GPT-5 mini ↓1 64.0
- 5 GPT-4.1 ↓1 63.0
- 6 Claude 3.7 Sonnet ↑10 58.0
- 7
Gemini 2.5 Flash Jun 2025 ↓2 55.0
- 8 Claude 3.5 Sonnet (October 2024) ↑10 53.0
- 9
Gemini 2.0 Flash Dec 2024 45.0
- 10
Qwen2.5-VL-72B 43.0
outlier
51 models we track have a score here.
- 1 GPT-5 78.0
- 2 Claude 3.7 Sonnet ↑14 68.0
- 3 Claude 3.5 Sonnet (October 2024) ↑15 63.0
- 4
Gemini 1.5 Pro Sep 2024 ↑9 58.0
- 5 GPT-4.1 ↓1 57.0
- 6
Seed1.5-VL ↓4 55.0
- 7
Gemini 1.5 Flash Sep 2024 ↑14 51.0
- 8
Qwen2.5-VL-72B ↑2 43.0
- 9
Qwen2.5-VL-32B ↑3 43.0
- 10
Gemini 2.5 Flash Jun 2025 ↓5 41.0
position
51 models we track have a score here.
- 1 GPT-4.1 ↑3 69.0
- 2
Seed1.5-VL 62.0
- 3
Gemini 2.5 Flash Jun 2025 ↑2 62.0
- 4 GPT-5 ↓3 62.0
- 5 GPT-5 mini ↓2 61.0
- 6
Gemini 2.0 Flash Dec 2024 ↑3 55.0
- 7
Qwen2.5-VL-72B ↑3 52.0
- 8 MiMo-VL-7B-RL ↑6 47.0
- 9 GPT-4.1 mini ↑2 46.0
- 10
Qwen2.5-VL-32B ↑2 45.0
rotation
51 models we track have a score here.
- 1 GPT-5 63.0
- 2 GPT-5 mini ↑1 59.0
- 3
Gemini 2.5 Flash Jun 2025 ↑2 51.0
- 4 Claude 3.5 Sonnet (October 2024) ↑14 47.0
- 5
Gemini 2.0 Flash Dec 2024 ↑4 47.0
- 6
Seed1.5-VL ↓4 42.0
- 7
Gemini 1.5 Pro Sep 2024 ↑6 39.0
- 8
Gemini 2.5 Pro Mar 2025 ↑18 39.0
- 9
InternVL3-38B ↓2 39.0
- 10 Qwen2-VL-72B-Instruct ↓4 38.0
rotation_status
51 models we track have a score here.
- 1
InternVL2.5-8B ↑30 62.0
- 2
Seed1.5-VL 62.0
- 3
Gemini 1.5 Pro Sep 2024 ↑10 61.0
- 4
DeepSeek-VL2-Small ↑34 61.0
- 5
Gemini 2.5 Pro Mar 2025 ↑21 60.0
- 6
LLaVA-OneVision 7B ↑33 59.0
- 7 Claude 3.5 Sonnet (October 2024) ↑11 59.0
- 8 Kimi-VL-A3B-Instruct ↑20 59.0
- 9
InternVL3-38B ↓2 59.0
- 10
Gemma 3 4B ↑36 58.0
height_higher
51 models we track have a score here.
- 1 GPT-5 71.8
- 2
Gemini 2.5 Flash Jun 2025 ↑3 68.8
- 3 GPT-5 mini 68.5
- 4 GPT-5 nano ↑16 66.5
- 5 MiMo-VL-7B-RL ↑9 65.1
- 6
InternVL3-78B ↑2 63.3
- 7 LLaVA-OneVision 72B ↑16 63.1
- 8 GPT-4.1 mini ↑3 62.4
- 9
Gemini 2.5 Pro Mar 2025 ↑17 62.3
- 10
Seed1.5-VL ↓8 62.2
location_above
51 models we track have a score here.
- 1 MiMo-VL-7B-RL ↑13 75.7
- 2
Seed1.5-VL 74.6
- 3 GPT-5 ↓2 72.3
- 4 GPT-5 mini ↓1 71.3
- 5
Gemini 2.5 Flash Jun 2025 71.0
- 6
SAIL-VL-1.6-8B ↑23 70.2
- 7
InternVL3-78B ↑1 69.9
- 8 GPT-5 nano ↑12 68.6
- 9
Ovis2-16B ↑13 68.5
- 10 Qwen2-VL-72B-Instruct ↓4 67.7
location_closer_to_camera
51 models we track have a score here.
- 1
InternVL3-78B ↑7 88.8
- 2
Seed1.5-VL 87.5
- 3
InternVL3-38B ↑4 87.4
- 4
InternVL3-14B ↑13 87.2
- 5 GPT-5 ↓4 85.3
- 6 GPT-5 mini ↓3 85.2
- 7
InternVL3-8B ↑17 85.0
- 8
Ovis2-34B ↑11 84.9
- 9
InternVL2_5-78B ↑6 84.1
- 10 GPT-4.1 ↓6 83.7
location_next_to
51 models we track have a score here.
- 1
Seed1.5-VL ↑1 83.5
- 2
Gemini 2.5 Flash Jun 2025 ↑3 79.6
- 3 GPT-5 ↓2 79.6
- 4 GPT-4.1 79.5
- 5 MiMo-VL-7B-RL ↑9 78.9
- 6 GPT-4.1 mini ↑5 78.0
- 7 GPT-5 nano ↑13 75.2
- 8
Gemini 2.5 Pro Mar 2025 ↑18 74.9
- 9
InternVL2_5-78B ↑6 74.8
- 10
Ovis2-34B ↑9 73.9
multi_object_closer_to
51 models we track have a score here.
- 1 GPT-5 mini ↑2 74.7
- 2 GPT-5 ↓1 73.0
- 3
Gemini 2.5 Flash Jun 2025 ↑2 71.9
- 4
Seed1.5-VL ↓2 71.6
- 5
InternVL3-78B ↑3 70.7
- 6
InternVL3-14B ↑11 70.1
- 7
InternVL3-38B 69.4
- 8 GPT-4.1 ↓4 69.0
- 9 GPT-4.1 mini ↑2 68.4
- 10 GPT-5 nano ↑10 68.3
multi_object_facing
51 models we track have a score here.
- 1
Seed1.5-VL ↑1 72.7
- 2
Gemini 2.5 Flash Jun 2025 ↑3 71.8
- 3 GPT-5 ↓2 71.7
- 4 GPT-5 mini ↓1 71.7
- 5
InternVL3-38B ↑2 71.2
- 6
Ovis2-16B ↑16 71.1
- 7 Qwen2-VL-72B-Instruct ↓1 69.5
- 8 GPT-5 nano ↑12 69.4
- 9
Ovis2-8B ↑21 68.8
- 10 MiMo-VL-7B-RL ↑4 68.4
multi_object_parallel
51 models we track have a score here.
- 1 GPT-5 66.5
- 2 GPT-4.1 ↑2 64.3
- 3 GPT-5 mini 62.5
- 4
InternVL3-78B ↑4 62.4
- 5 MiniCPM-o-2_6 ↑32 61.5
- 6 GPT-5 nano ↑14 61.1
- 7
Ovis2-34B ↑12 60.9
- 8
InternVL3-38B ↓1 60.5
- 9
Ovis2-16B ↑13 60.2
- 10 GPT-4.1 mini ↑1 60.0
multi_object_same_direction
51 models we track have a score here.
- 1
Seed1.5-VL ↑1 61.2
- 2 GPT-5 ↓1 61.0
- 3 GPT-5 mini 58.9
- 4
Qwen2.5-VL-32B ↑8 58.4
- 5 GPT-5 nano ↑15 58.1
- 6 Qwen2-VL-72B-Instruct 57.8
- 7 GPT-4.1 mini ↑4 56.8
- 8
InternVL3-38B ↓1 56.5
- 9
Ovis2-34B ↑10 56.2
- 10
InternVL3-2B ↑33 55.7
multi_object_viewpoint_towards_object
51 models we track have a score here.
- 1 GPT-5 40.5
- 2
Seed1.5-VL 39.2
- 3 GPT-5 mini 37.3
- 4 GPT-5 nano ↑16 36.1
- 5 GPT-4.1 mini ↑6 35.1
- 6
InternVL3-78B ↑2 34.3
- 7 MiMo-VL-7B-RL ↑7 33.2
- 8 GPT-4.1 ↓4 32.0
- 9
Gemini 2.5 Flash Jun 2025 ↓4 32.0
- 10 LLaVA-OneVision 72B ↑13 31.3
orientation_in_front_of
51 models we track have a score here.
- 1 GPT-5 81.0
- 2 GPT-5 mini ↑1 78.9
- 3
Seed1.5-VL ↓1 71.9
- 4
InternVL3-78B ↑4 68.5
- 5 GPT-5 nano ↑15 68.3
- 6
Gemini 2.5 Flash Jun 2025 ↓1 68.2
- 7
Qwen2.5-VL-72B ↑3 68.0
- 8
InternVL2.5-8B ↑23 66.9
- 9 Claude 3.7 Sonnet ↑7 66.1
- 10 Qwen2-VL-72B-Instruct ↓4 66.0
orientation_on_the_left
51 models we track have a score here.
- 1 GPT-5 nano ↑19 53.9
- 2 GPT-5 ↓1 52.4
- 3
Gemma 3 12B ↑31 45.0
- 4
Gemma 3 4B ↑42 44.7
- 5 GPT-4.1 nano ↑36 44.3
- 6
Gemini 2.0 Flash Dec 2024 ↑3 44.0
- 7
SmolVLM2-2.2B-Instruct ↑42 43.6
- 8
InternVL3-1B ↑37 43.6
- 9 GPT-5 mini ↓6 43.3
- 10 Phi-3.5-vision-instruct ↑34 42.7
orientation_viewpoint
51 models we track have a score here.
- 1 GPT-5 54.8
- 2 GPT-4.1 mini ↑9 54.2
- 3
Seed1.5-VL ↓1 53.6
- 4
Qwen2.5-VL-72B ↑6 52.5
- 5 GPT-5 mini ↓2 52.3
- 6
InternVL3-38B ↑1 52.1
- 7
InternVL3-78B ↑1 51.7
- 8
SAIL-VL-1.6-8B ↑21 50.0
- 9
Gemini 1.5 Pro Sep 2024 ↑4 49.1
- 10 Qwen2-VL-72B-Instruct ↓4 49.0
L1_single
50 models we track have a score here.
- 1 Qwen2-VL-72B-Instruct ↑5 83.0
- 2
InternVL3-38B ↑5 81.5
- 3 GPT-5 ↓2 80.7
- 4
Ovis2-34B ↑15 80.3
- 5
InternVL3-78B ↑3 77.0
- 6 Qwen2-VL-7B-Instruct ↑19 76.7
- 7
Ovis2-16B ↑15 76.7
- 8
Seed1.5-VL ↓6 76.4
- 9
InternVL3-14B ↑8 76.1
- 10
Ovis2-8B ↑20 75.4
L2_objects
50 models we track have a score here.
- 1 Qwen2-VL-72B-Instruct ↑5 82.7
- 2
InternVL3-38B ↑5 81.1
- 3 Qwen2-VL-7B-Instruct ↑22 78.8
- 4
InternVL3-14B ↑13 77.1
- 5
InternVL3-78B ↑3 76.5
- 6 GPT-5 ↓5 74.7
- 7
Ovis2-34B ↑12 73.5
- 8
InternVL3-8B ↑16 72.9
- 9 GPT-5 mini ↓6 70.6
- 10 GPT-4.1 ↓6 69.2
L3_2d_spatial
50 models we track have a score here.
- 1 Qwen2-VL-72B-Instruct ↑5 83.7
- 2
InternVL3-38B ↑5 81.3
- 3 Qwen2-VL-7B-Instruct ↑22 80.3
- 4
InternVL3-14B ↑13 79.0
- 5
InternVL3-78B ↑3 76.8
- 6
InternVL3-8B ↑18 73.8
- 7 GPT-5 ↓6 73.2
- 8
Ovis2-34B ↑11 70.8
- 9 GPT-5 mini ↓6 66.0
- 10
Qwen2.5-VL-32B ↑2 65.6
L4_occ
50 models we track have a score here.
- 1 Qwen2-VL-72B-Instruct ↑5 66.0
- 2 GPT-5 ↓1 62.5
- 3
Seed1.5-VL ↓1 61.4
- 4
InternVL3-38B ↑3 58.5
- 5
InternVL3-78B ↑3 57.4
- 6 GPT-5 mini ↓3 55.5
- 7 GPT-4.1 ↓3 55.3
- 8 Qwen2-VL-7B-Instruct ↑17 55.0
- 9 MiMo-VL-7B-RL ↑5 54.4
- 10
InternVL3-14B ↑7 54.3
L4_pose
50 models we track have a score here.
- 1 GPT-5 48.9
- 2
Seed1.5-VL 48.6
- 3 Qwen2-VL-72B-Instruct ↑3 48.1
- 4 GPT-5 mini ↓1 46.1
- 5
Ovis2-16B ↑17 46.0
- 6 GPT-4.1 ↓2 46.0
- 7
InternVL3-38B 45.8
- 8
Ovis2-34B ↑11 45.7
- 9
Gemini 2.0 Flash Dec 2024 45.6
- 10
Gemini 1.5 Pro Sep 2024 ↑3 44.6
L5_6d_spatial
50 models we track have a score here.
- 1 Qwen2-VL-72B-Instruct ↑5 49.5
- 2
Ovis2-34B ↑17 49.4
- 3
InternVL3-38B ↑4 47.7
- 4 Qwen2-VL-7B-Instruct ↑21 47.1
- 5
InternVL3-14B ↑12 46.7
- 6
InternVL3-8B ↑18 45.7
- 7 GPT-5 mini ↓4 45.7
- 8 GPT-5 ↓7 45.3
- 9 LLaVA-OneVision 72B ↑14 45.1
- 10 GPT-4.1 ↓6 44.9
L5_collision
50 models we track have a score here.
- 1 Qwen2-VL-72B-Instruct ↑5 54.1
- 2
Ovis2-34B ↑17 51.9
- 3 Qwen2-VL-7B-Instruct ↑22 50.6
- 4
Seed1.5-VL ↓2 50.4
- 5 GPT-5 ↓4 50.4
- 6
InternVL3-38B ↑1 49.7
- 7 GPT-5 mini ↓4 49.5
- 8
Ovis2-16B ↑14 49.2
- 9 GPT-4.1 ↓5 49.1
- 10
InternLM-XComposer2.5-7B ↑26 48.1
Attr-Appr
51 models we track have a score here.
- 1 GPT-5 mini ↑2 39.4
- 2
Phi-4-multimodal ↑33 37.9
- 3
Gemini 2.5 Flash Jun 2025 ↑2 37.9
- 4
Qwen2.5-VL-32B ↑8 36.4
- 5 GPT-5 ↓4 36.4
- 6 Phi-3.5-vision-instruct ↑38 31.8
- 7
Ovis2-34B ↑12 31.8
- 8 Kimi-VL-A3B-Instruct ↑20 31.8
- 9
Idefics3-8B-Llama3 ↑38 30.3
- 10
SmolVLM2-2.2B-Instruct ↑39 30.3
Attr-Meas
51 models we track have a score here.
- 1
Gemini 2.5 Flash Jun 2025 ↑4 60.9
- 2 GPT-5 ↓1 60.9
- 3 GPT-4.1 ↑1 46.9
- 4 GPT-5 mini ↓1 46.9
- 5
Gemini 1.5 Flash Sep 2024 ↑16 45.3
- 6
Gemini 2.0 Flash Dec 2024 ↑3 45.3
- 7
Gemini 2.5 Pro Mar 2025 ↑19 43.8
- 8
Qwen2.5-VL-72B ↑2 42.2
- 9
InternVL3-38B ↓2 42.2
- 10 Qwen2-VL-72B-Instruct ↓4 40.6
MSR
51 models we track have a score here.
- 1 GPT-5 42.4
- 2
Qwen2.5-VL-72B ↑8 33.3
- 3 Qwen2-VL-7B-Instruct ↑22 32.8
- 4
Ovis2-16B ↑18 32.8
- 5 MiMo-VL-7B-RL ↑9 32.8
- 6 GPT-5 mini ↓3 32.8
- 7
Gemini 1.5 Flash Sep 2024 ↑14 31.8
- 8
Ovis2-34B ↑11 31.3
- 9
InternVL2.5-8B ↑22 30.8
- 10
InternVL3-8B ↑14 30.8
Motion-Cam
51 models we track have a score here.
- 1 GPT-4.1 ↑3 37.8
- 2
Seed1.5-VL 37.8
- 3
Qwen2.5-VL-72B ↑7 33.8
- 4
InternVL3-78B ↑4 33.8
- 5 Qwen2-VL-72B-Instruct ↑1 32.4
- 6 GPT-4.1 mini ↑5 32.4
- 7 GPT-5 ↓6 32.4
- 8
Idefics3-8B-Llama3 ↑39 31.1
- 9
Janus-Pro-7B ↑39 28.4
- 10 MiMo-VL-7B-RL ↑4 28.4
Motion-Obj
51 models we track have a score here.
- 1
Gemini 2.5 Flash Jun 2025 ↑4 38.2
- 2
Gemini 1.5 Pro Sep 2024 ↑11 36.8
- 3 Claude 3.5 Sonnet (October 2024) ↑15 36.8
- 4 GPT-5 ↓3 36.8
- 5 Qwen2-VL-72B-Instruct ↑1 35.5
- 6 MiniCPM-o-2_6 ↑31 35.5
- 7 GPT-4.1 ↓3 35.5
- 8
SAIL-VL-1.6-8B ↑21 35.5
- 9
InternVL2_5-78B ↑6 34.2
- 10
Janus-Pro-7B ↑38 34.2
Pos-Cam-Cam
51 models we track have a score here.
- 1 GPT-5 43.0
- 2 LLaVA-OneVision 72B ↑21 40.9
- 3 Qwen2-VL-72B-Instruct ↑3 40.9
- 4 GPT-5 mini ↓1 38.7
- 5 Claude 3.5 Sonnet (October 2024) ↑13 35.5
- 6
InternVL2_5-78B ↑9 35.5
- 7 GPT-4.1 nano ↑34 35.5
- 8
Seed1.5-VL ↓6 35.5
- 9 MiMo-VL-7B-RL ↑5 35.5
- 10
Gemini 2.5 Flash Jun 2025 ↓5 35.5
Pos-Cam-Obj
51 models we track have a score here.
- 1 GPT-5 48.8
- 2
Ovis2-8B ↑28 40.7
- 3
SAIL-VL-1.6-8B ↑26 38.4
- 4
DeepSeek-VL2 ↑29 36.0
- 5
InternVL2.5-8B ↑26 34.9
- 6
Ovis2-16B ↑16 34.9
- 7 GPT-4.1 nano ↑34 34.9
- 8
Gemini 2.0 Flash Dec 2024 ↑1 33.7
- 9
Gemma 3 12B ↑25 33.7
- 10
DeepSeek-VL2-Small ↑28 32.6
Pos-Cam-Reg
51 models we track have a score here.
- 1 GPT-5 51.8
- 2 GPT-5 mini ↑1 49.4
- 3 GPT-4.1 nano ↑38 47.0
- 4 Claude 3.5 Sonnet (October 2024) ↑14 45.8
- 5
InternVL2_5-78B ↑10 42.2
- 6 Claude 3.7 Sonnet ↑10 41.0
- 7
Gemini 2.5 Flash Jun 2025 ↓2 41.0
- 8
Gemini 2.0 Flash Dec 2024 ↑1 38.6
- 9 Qwen2-VL-72B-Instruct ↓3 37.3
- 10
InternVL3-8B ↑14 37.3
Pos-Obj-Obj
51 models we track have a score here.
- 1
Janus-Pro-7B ↑47 40.4
- 2
SmolVLM2-2.2B-Instruct ↑47 35.1
- 3
InternVL3-1B ↑42 35.1
- 4 GPT-5 ↓3 35.1
- 5 GPT-5 nano ↑15 35.1
- 6 Phi-3.5-vision-instruct ↑38 34.0
- 7 Qwen2-VL-7B-Instruct ↑18 34.0
- 8
LLaVA-OneVision 7B ↑31 33.0
- 9
Pixtral 12B ↑31 33.0
- 10
DeepSeek-VL2-Small ↑28 31.9
Pos-Obj-Reg
51 models we track have a score here.
- 1 MiMo-VL-7B-RL ↑13 45.9
- 2
Ovis2-34B ↑17 44.7
- 3
InternVL3-8B ↑21 43.5
- 4 GPT-5 ↓3 42.4
- 5
Gemini 1.5 Flash Sep 2024 ↑16 41.2
- 6
Qwen2.5-VL-72B ↑4 41.2
- 7 Claude 3.5 Sonnet (October 2024) ↑11 40.0
- 8
InternVL2_5-78B ↑7 40.0
- 9
Ovis2-8B ↑21 40.0
- 10
Seed1.5-VL ↓8 40.0
Pos-Reg-Reg
51 models we track have a score here.
- 1
Ovis2-16B ↑21 40.7
- 2 Claude 3.5 Sonnet (October 2024) ↑16 38.3
- 3
InternVL2_5-78B ↑12 38.3
- 4
Qwen2.5-VL-72B ↑6 38.3
- 5
Gemini 2.0 Flash Dec 2024 ↑4 34.6
- 6
InternVL3-2B ↑37 34.6
- 7 GPT-4.1 ↓3 34.6
- 8
Gemini 1.5 Flash Sep 2024 ↑13 33.3
- 9 LLaVA-OneVision 72B ↑14 32.1
- 10 Qwen2-VL-72B-Instruct ↓4 32.1
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: Official OpenLMM Spatial Leaderboard leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| GPT-5 | Standard | 62.5 | |||
| Standard | 57.3 | ||||
| GPT-5 mini | Standard | 57.1 | |||
| GPT-4.1 | Standard | 55.2 | |||
| Standard | 54.3 | ||||
| Qwen2-VL-72B-Instruct | Standard | 53.1 | |||
| Standard | 52.8 | ||||
| Standard | 52.8 | ||||
| Standard | 51.8 | ||||
| Standard | 51.4 | ||||
| GPT-4.1 mini | Standard | 51.1 | |||
| Standard | 51.0 | ||||
| Standard | 50.8 | ||||
| MiMo-VL-7B-RL | Standard | 50.5 | |||
| Standard | 49.2 | ||||
| Claude 3.7 Sonnet | Standard | 49.1 | |||
| Standard | 49.1 | ||||
| Claude 3.5 Sonnet (October 2024) | Standard | 49.0 | |||
| Standard | 48.9 | ||||
| GPT-5 nano | Standard | 48.2 | |||
| Standard | 47.6 | ||||
| Standard | 47.5 | ||||
| GPT-5 | Circular | 47.5 | |||
| LLaVA-OneVision 72B | Standard | 47.2 | |||
| Standard | 46.7 | ||||
| Qwen2-VL-7B-Instruct | Standard | 45.9 | |||
| Circular | 45.7 | ||||
| Circular | 45.0 | ||||
| Standard | 44.1 | ||||
| Kimi-VL-A3B-Instruct | Standard | 44.0 | |||
| Standard | 43.7 | ||||
| Standard | 43.5 | ||||
| Standard | 43.3 | ||||
| Standard | 43.3 | ||||
| Standard | 42.7 | ||||
| Standard | 42.3 | ||||
| Standard | 42.1 | ||||
| Standard | 41.7 | ||||
| MiniCPM-o-2_6 | Standard | 41.5 | |||
| Circular | 41.3 | ||||
| Circular | 41.2 | ||||
| Standard | 40.9 | ||||
| Standard | 40.8 | ||||
| GPT-4.1 | Circular | 40.5 | |||
| GPT-5 mini | Circular | 40.0 | |||
| Standard | 39.5 | ||||
| Circular | 39.3 | ||||
| GPT-4.1 nano | Standard | 39.0 | |||
| Standard | 38.5 | ||||
| Standard | 38.4 | ||||
| Phi-3.5-vision-instruct | Standard | 38.2 | |||
| Standard | 38.1 | ||||
| Circular | 37.6 | ||||
| Standard | 37.4 | ||||
| Circular | 36.9 | ||||
| Qwen2-VL-72B-Instruct | Circular | 35.9 | |||
| Circular | 35.9 | ||||
| Circular | 35.9 | ||||
| GPT-4.1 mini | Circular | 34.9 | |||
| Circular | 34.8 | ||||
| Circular | 34.6 | ||||
| Circular | 34.0 | ||||
| LLaVA-OneVision 72B | Circular | 33.9 | |||
| Claude 3.5 Sonnet (October 2024) | Circular | 33.4 | |||
| Circular | 33.3 | ||||
| Standard | 33.1 | ||||
| MiMo-VL-7B-RL | Circular | 33.0 | |||
| Claude 3.7 Sonnet | Circular | 32.5 | |||
| Standard | 32.4 | ||||
| Circular | 31.9 | ||||
| Circular | 31.3 | ||||
| Circular | 30.5 | ||||
| Standard | 30.1 | ||||
| MiniCPM-o-2_6 | Circular | 29.8 | |||
| Kimi-VL-A3B-Instruct | Circular | 29.8 | |||
| Circular | 29.7 | ||||
| Circular | 29.6 | ||||
| Circular | 29.4 | ||||
| GPT-5 nano | Circular | 29.1 | |||
| Circular | 28.3 | ||||
| Circular | 28.2 | ||||
| Circular | 28.1 | ||||
| Circular | 27.2 | ||||
| Circular | 26.8 | ||||
| Circular | 26.7 | ||||
| Qwen2-VL-7B-Instruct | Circular | 25.8 | |||
| Phi-3.5-vision-instruct | Circular | 24.7 | |||
| Circular | 23.7 | ||||
| Circular | 23.3 | ||||
| Circular | 23.1 | ||||
| Circular | 21.7 | ||||
| Circular | 20.1 | ||||
| Circular | 20.0 | ||||
| llava-v1.6-vicuna-7b | Standard | 19.9 | |||
| llava-v1.6-vicuna-13b | Standard | 19.6 | |||
| Circular | 19.1 | ||||
| GPT-4.1 nano | Circular | 18.8 | |||
| Circular | 15.7 | ||||
| Circular | 15.2 | ||||
| llava-v1.6-vicuna-7b | Circular | 14.1 | |||
| llava-v1.6-vicuna-13b | Circular | 12.7 |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: Official OpenLMM Spatial Leaderboard leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| GPT-5 | Standard | 62.5 | |||
| Standard | 57.3 | ||||
| GPT-5 mini | Standard | 57.1 | |||
| GPT-4.1 | Standard | 55.2 | |||
| Standard | 54.3 | ||||
| Qwen2-VL-72B-Instruct | Standard | 53.1 | |||
| Standard | 52.8 | ||||
| Standard | 52.8 | ||||
| Standard | 51.8 | ||||
| Standard | 51.4 | ||||
| GPT-4.1 mini | Standard | 51.1 | |||
| Standard | 51.0 | ||||
| Standard | 50.8 | ||||
| MiMo-VL-7B-RL | Standard | 50.5 | |||
| Standard | 49.2 | ||||
| Claude 3.7 Sonnet | Standard | 49.1 | |||
| Standard | 49.1 | ||||
| Claude 3.5 Sonnet (October 2024) | Standard | 49.0 | |||
| Standard | 48.9 | ||||
| GPT-5 nano | Standard | 48.2 | |||
| Standard | 47.6 | ||||
| Standard | 47.5 | ||||
| GPT-5 | Circular | 47.5 | |||
| LLaVA-OneVision 72B | Standard | 47.2 | |||
| Standard | 46.7 | ||||
| Qwen2-VL-7B-Instruct | Standard | 45.9 | |||
| Circular | 45.7 | ||||
| Circular | 45.0 | ||||
| Standard | 44.1 | ||||
| Kimi-VL-A3B-Instruct | Standard | 44.0 | |||
| Standard | 43.7 | ||||
| Standard | 43.5 | ||||
| Standard | 43.3 | ||||
| Standard | 43.3 | ||||
| Standard | 42.7 | ||||
| Standard | 42.3 | ||||
| Standard | 42.1 | ||||
| Standard | 41.7 | ||||
| MiniCPM-o-2_6 | Standard | 41.5 | |||
| Circular | 41.3 | ||||
| Circular | 41.2 | ||||
| Standard | 40.9 | ||||
| Standard | 40.8 | ||||
| GPT-4.1 | Circular | 40.5 | |||
| GPT-5 mini | Circular | 40.0 | |||
| Standard | 39.5 | ||||
| Circular | 39.3 | ||||
| GPT-4.1 nano | Standard | 39.0 | |||
| Standard | 38.5 | ||||
| Standard | 38.4 | ||||
| Phi-3.5-vision-instruct | Standard | 38.2 | |||
| Standard | 38.1 | ||||
| Circular | 37.6 | ||||
| Standard | 37.4 | ||||
| Circular | 36.9 | ||||
| Qwen2-VL-72B-Instruct | Circular | 35.9 | |||
| Circular | 35.9 | ||||
| Circular | 35.9 | ||||
| GPT-4.1 mini | Circular | 34.9 | |||
| Circular | 34.8 | ||||
| Circular | 34.6 | ||||
| Circular | 34.0 | ||||
| LLaVA-OneVision 72B | Circular | 33.9 | |||
| Claude 3.5 Sonnet (October 2024) | Circular | 33.4 | |||
| Circular | 33.3 | ||||
| Standard | 33.1 | ||||
| MiMo-VL-7B-RL | Circular | 33.0 | |||
| Claude 3.7 Sonnet | Circular | 32.5 | |||
| Standard | 32.4 | ||||
| Circular | 31.9 | ||||
| Circular | 31.3 | ||||
| Circular | 30.5 | ||||
| Standard | 30.1 | ||||
| MiniCPM-o-2_6 | Circular | 29.8 | |||
| Kimi-VL-A3B-Instruct | Circular | 29.8 | |||
| Circular | 29.7 | ||||
| Circular | 29.6 | ||||
| Circular | 29.4 | ||||
| GPT-5 nano | Circular | 29.1 | |||
| Circular | 28.3 | ||||
| Circular | 28.2 | ||||
| Circular | 28.1 | ||||
| Circular | 27.2 | ||||
| Circular | 26.8 | ||||
| Circular | 26.7 | ||||
| Qwen2-VL-7B-Instruct | Circular | 25.8 | |||
| Phi-3.5-vision-instruct | Circular | 24.7 | |||
| Circular | 23.7 | ||||
| Circular | 23.3 | ||||
| Circular | 23.1 | ||||
| Circular | 21.7 | ||||
| Circular | 20.1 | ||||
| Circular | 20.0 | ||||
| llava-v1.6-vicuna-7b | Standard | 19.9 | |||
| llava-v1.6-vicuna-13b | Standard | 19.6 | |||
| Circular | 19.1 | ||||
| GPT-4.1 nano | Circular | 18.8 | |||
| Circular | 15.7 | ||||
| Circular | 15.2 | ||||
| llava-v1.6-vicuna-7b | Circular | 14.1 | |||
| llava-v1.6-vicuna-13b | Circular | 12.7 |
2 models have held number one on the official leaderboard since its release in June 2025
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official OpenLMM Spatial Leaderboard leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official OpenLMM Spatial Leaderboard 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 Other
- Artificial Analysis Intelligence Index v5 In use Version 5 of Artificial Analysis's index: nine evaluations in four equal categories, most with private test sets, run three times each in its own agent harness. Claude Fable 5.2 64.6
- Personal Superintelligence Index (Meta) In use Meta's composite of seven personal and knowledge-work evals: Agents' Last Exam, AutomationBench-AA, FocusBench, JobBench, GDP.pdf, VBT and DRACO.
- Vals Science Index In use Agents are scored on the equal average of BioMysteryBench, MysteryMechanism and Terminal-Bench-Science. GPT-6.2 Astra 68.4%
- Decision Index Public v0.3 In use Decision models return probabilities over labeled options in public tasks covering knowledge, language, retrieval, tools, and arts. A chance-corrected weighted aggregate combines the task scores. Granite Switch 4.1 30B 51.3
- SAGE v1.1 (Vals) In use Models grade handwritten undergraduate math and computer science work against rubrics. Scored by class-balanced accuracy of each rubric check against expert grading. Claude Opus 5.5 91.3%
- MTEB Fading Massive Text Embedding Benchmark: retrieval, clustering, classification across languages and modalities.
Qwen3-Embedding-8B 72.2%
Data audit
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 15 months ago, in August 2025, for GPT-5. 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 (59 entries).
- Pass Models scored 51 models we track, plus 8 other entries on its leaderboard.
- Warning Newest score The newest came 15 months ago, in August 2025, for GPT-5.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5 (#71).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on huggingface.co: 101 rows for models we track and 8 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.