Individual benchmark scores plotted by date.
| Organisation | Model | Reported | Top Score | Info | Self Reported | Source |
|---|---|---|---|---|---|---|
| GPT 5.4 | 05 Mar 2026 | 82.10% | With tools | Yes | Source | |
| Gemini 3 Flash Preview | 17 Dec 2025 | 81.20% | - | Yes | Source | |
| GPT 5 Pro | 07 Aug 2025 | 81.20% | inferred family alias from gpt-5.4 (score=0.4083; benches=19) | Yes | Source | |
| GPT 5 Search API | 14 Oct 2025 | 81.20% | inferred family alias from gpt-5.4 (score=0.3050; benches=19) | Yes | Source | |
| Gemini 3 Pro Preview | 18 Nov 2025 | 81% | - | Yes | Source | |
| Gemini 3 Pro Image Preview (Nano Banana Pro) | 20 Nov 2025 | 81% | inferred modality/version alias from gemini-3-pro-preview | Yes | Source | |
| Gemini 3.1 Pro Preview | 19 Feb 2026 | 80.50% | Search + Python + Browse | Yes | - | |
| Muse Spark | 08 Apr 2026 | 80.40% | Multimodal understanding | Yes | Source | |
| GPT 5.2 Chat | 11 Dec 2025 | 79.50% | inferred alias from gpt-5.2-2025-12-11 | Yes | Source | |
| Qwen 3.5 397B A17B | 16 Feb 2026 | 79% | - | Yes | Source | |
| Qwen 3.6 Plus | 01 Apr 2026 | 78.80% | - | Yes | Source | |
| Kimi K2.5 | 27 Jan 2026 | 78.50% | - | Yes | Source | |
| GPT 5 | 07 Aug 2025 | 78.40% | With Thinking, Pass @ 1 | Yes | Source | |
| Seed 2.0 Pro | 14 Feb 2026 | 78.20% | Seed2 official benchmark table | MMMU-Pro | Yes | Source | |
| GPT 5.4 Mini | 17 Mar 2026 | 78% | With Python | Yes | Source | |
| Claude Opus 4.6 | 05 Feb 2026 | 77.30% | With tools | Yes | Source | |
| Gemma 4 31B | 02 Apr 2026 | 76.90% | - | Yes | Source | |
| Qwen 3.5 122B A10B | 24 Feb 2026 | 76.90% | - | Yes | Source | |
| Gemini 3.1 Flash Lite Preview | 03 Mar 2026 | 76.80% | - | Yes | Source | |
| Gemini 3.1 Flash Image Preview (Nano Banana 2) | 26 Feb 2026 | 76.80% | inferred modality/version alias from gemini-3.1-flash-lite-preview | Yes | Source | |
| Veo 3.1 Lite Preview | 31 Mar 2026 | 76.80% | manual fallback alias from gemini-3.1-flash-lite-preview | Yes | Source | |
| Seed 2.0 Lite | 14 Feb 2026 | 76% | Seed2 official benchmark table | MMMU-Pro | Yes | Source | |
| Claude Sonnet 4.6 | 17 Feb 2026 | 75.60% | With tools | Yes | Source | |
| Qwen 3.5 35B A3B | 24 Feb 2026 | 75.10% | - | Yes | Source | |
| Qwen 3.5 27B | 24 Feb 2026 | 75% | - | Yes | Source | |
| Qwen 3.5 Flash | 23 Feb 2026 | 75% | inferred family alias from qwen3.5-27b (score=0.4147; benches=81) | Yes | Source | |
| GPT 5 Mini | 07 Aug 2025 | 74.10% | High Reasoning Effort | Yes | Source | |
| Gemma 4 26B A4B | 02 Apr 2026 | 73.80% | - | Yes | Source | |
| Seed 2.0 Mini | 14 Feb 2026 | 71.40% | Seed2 official benchmark table | MMMU-Pro | Yes | Source | |
| GPT 5.4 Nano | 17 Mar 2026 | 69.50% | With Python | Yes | Source | |
| Qwen 3 VL 235B A22B Thinking | - | 69.30% | - | Yes | - | |
| Qwen 3 VL 235B A22B Instruct | - | 68.10% | - | Yes | - | |
| Qwen 3 VL 32B Thinking | - | 68.10% | - | Yes | - | |
| Qwen 3 VL 32B Instruct | - | 65.30% | - | Yes | - | |
| Nova 2 Pro | 02 Dec 2025 | 63.50% | - | Yes | Source | |
| Qwen 3 VL 30B A3B Thinking | - | 63% | - | Yes | - | |
| GPT 5 Nano | 07 Aug 2025 | 62.60% | High Reasoning Effort | Yes | Source | |
| Nova 2 Lite | 02 Dec 2025 | 61.80% | - | Yes | Source | |
| Qwen 3 VL 30B A3B Instruct | - | 60.40% | - | Yes | - | |
| Qwen 3 Reranker 8B | - | 60.40% | inferred family alias from qwen3-vl-8b-thinking (score=0.3850; benches=50) | Yes | - | |
| Qwen 3 VL 8B Thinking | - | 60.40% | - | Yes | - | |
| Qwen 3 VL Reranker 8B | - | 60.40% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5275; benches=50) | Yes | - | |
| Qwen 3 8B | - | 60.40% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50) | Yes | - | |
| Qwen 3 Embedding 8B | - | 60.40% | inferred family alias from qwen3-vl-8b-thinking (score=0.3850; benches=50) | Yes | - | |
| Qwen 3 Guard Stream 8B | - | 60.40% | inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50) | Yes | - | |
| Qwen 3 Guard Gen 8B | - | 60.40% | inferred family alias from qwen3-vl-8b-thinking (score=0.3400; benches=50) | Yes | - | |
| Qwen 3 VL Embedding 8B | - | 60.40% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5232; benches=50) | Yes | - | |
| Mistral Small 1.0 | 26 Feb 2024 | 60% | inferred family alias from mistral-small-latest (score=0.3650; benches=9) | Yes | Source | |
| Mistral Small Creative | 16 Dec 2025 | 60% | inferred family alias from mistral-small-latest (score=0.4273; benches=9) | Yes | Source | |
| Mistral Small 4 | 16 Mar 2026 | 60% | - | Yes | Source | |
| Mistral Small 2.0 | 17 Sept 2024 | 60% | inferred family alias from mistral-small-latest (score=0.3650; benches=9) | Yes | Source | |
| GPT 4o Transcribe | 20 Mar 2025 | 59.90% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Audio (2024-12-17) | 17 Dec 2024 | 59.90% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Transcribe Diarize | 15 Oct 2025 | 59.90% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Realtime Preview (2024-10-01) | 01 Oct 2024 | 59.90% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Search Preview | 11 Mar 2025 | 59.90% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Audio (2024-10-01) | 01 Oct 2024 | 59.90% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Audio (2025-06-03) | 03 Jun 2025 | 59.90% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o (2024-08-06) | 06 Aug 2024 | 59.90% | - | Yes | Source | |
| GPT 4o Realtime Preview (2025-06-03) | 03 Jun 2025 | 59.90% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| Llama 4 Maverick | 05 Apr 2025 | 59.60% | - | Yes | Source | |
| Qwen 3 4B Thinking 2507 | - | 57% | inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48) | Yes | - | |
| Qwen 3 4B Instruct 2507 | - | 57% | inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48) | Yes | - | |
| Qwen 3 Guard Gen 4B | - | 57% | inferred family alias from qwen3-vl-4b-thinking (score=0.3400; benches=48) | Yes | - | |
| Qwen 3 4B SafeRL | - | 57% | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | Yes | - | |
| Qwen 3 VL 4B Thinking | - | 57% | - | Yes | - | |
| Qwen 3 Guard Stream 4B | - | 57% | inferred family alias from qwen3-vl-4b-thinking (score=0.3371; benches=48) | Yes | - | |
| Qwen 3 Embedding 4B | - | 57% | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | Yes | - | |
| Qwen 3 Reranker 4B | - | 57% | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | Yes | - | |
| Qwen 3 VL 8B Instruct | - | 55.90% | - | Yes | - | |
| Qwen 3 VL 4B Instruct | - | 53.20% | - | Yes | - | |
| Qwen 2.5 Math 72B | - | 51.10% | inferred high-confidence family alias from qwen2.5-vl-72b (score=0.4700; benches=30) | Yes | Source | |
| Qwen 2.5 Math PRM 72B | - | 51.10% | inferred family alias from qwen2.5-vl-72b (score=0.4050; benches=30) | Yes | Source | |
| Qwen 2.5 Math RM 72B | - | 51.10% | inferred family alias from qwen2.5-vl-72b (score=0.4092; benches=30) | Yes | Source | |
| Qwen 2.5 VL 72B Instruct | - | 51.10% | - | Yes | Source | |
| Qwen 2.5 Coder 32B Instruct | - | 49.50% | inferred high-confidence family alias from qwen2.5-vl-32b (score=0.4641; benches=28) | Yes | Source | |
| Qwen 2.5 VL 3B Instruct | - | 49.50% | inferred high-confidence family alias from qwen2.5-vl-32b (score=0.4914; benches=28) | Yes | Source | |
| Qwen 2.5 VL 32B Instruct | - | 49.50% | - | Yes | Source | |
| Qwen 2 VL 72B | - | 46.20% | - | Yes | Source | |
| Qwen 2 VL 2B | - | 46.20% | inferred family alias from qwen2-vl-72b (score=0.4083; benches=15) | Yes | Source | |
| Llama 3.2 90B Vision Instruct | - | 45.20% | inferred high-confidence family alias from llama-3.2-90b-instruct (score=0.5873; benches=13) | Yes | Source | |
| Phi 4 multimodal instruct | 01 Feb 2025 | 38.50% | - | Yes | Source | |
| Qwen 2.5 1.5B | - | 38.30% | inferred family alias from qwen2.5-vl-7b (score=0.3131; benches=32) | Yes | Source | |
| Qwen 2.5 3B | - | 38.30% | inferred family alias from qwen2.5-vl-7b (score=0.3531; benches=32) | Yes | Source | |
| Qwen 2.5 0.5B | - | 38.30% | inferred family alias from qwen2.5-vl-7b (score=0.3131; benches=32) | Yes | Source | |
| Qwen 2.5 VL 7B Instruct | - | 38.30% | - | Yes | Source | |
| Qwen 2.5 Math 7B | - | 36.60% | inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4767; benches=45) | Yes | Source | |
| Qwen 2.5 Math 7B PRM800K | - | 36.60% | inferred family alias from qwen2.5-omni-7b (score=0.3696; benches=45) | Yes | Source | |
| Qwen 2.5 Coder 3B | - | 36.60% | inferred family alias from qwen2.5-omni-7b (score=0.3000; benches=45) | Yes | Source | |
| Qwen 2.5 Coder 7B | - | 36.60% | inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4700; benches=45) | Yes | Source | |
| Qwen 2.5 Math PRM 7B | - | 36.60% | inferred family alias from qwen2.5-omni-7b (score=0.4092; benches=45) | Yes | Source | |
| Qwen 2.5 Omni 7B | - | 36.60% | - | Yes | Source | |
| Qwen 2.5 Omni 3B | - | 36.60% | inferred high-confidence family alias from qwen2.5-omni-7b (score=0.4933; benches=45) | Yes | Source | |
| Llama 3.2 1B Instruct | 25 Sept 2024 | 33% | inferred family alias from llama-3.2-11b-instruct (score=0.4200; benches=11) | Yes | Source | |
| Llama 3.2 11B Vision Instruct | - | 33% | inferred high-confidence family alias from llama-3.2-11b-instruct (score=0.5873; benches=11) | Yes | Source |
