Individual benchmark scores plotted by date.
| Organisation | Model | Reported | Top Score | Info | Self Reported | Source |
|---|---|---|---|---|---|---|
| GPT 4.5 | 27 Feb 2025 | 90.80% | - | Yes | Source | |
| o1 preview | 12 Sept 2024 | 90.80% | - | Yes | Source | |
| Qwen 3 VL 235B A22B Thinking | - | 90.60% | - | Yes | - | |
| GPT OSS 120b | 05 Aug 2025 | 90% | High Reasoning Effort | Yes | Source | |
| GPT OSS Safeguard 120b | 29 Oct 2025 | 90% | inferred high-confidence family alias from gpt-oss-120b (score=0.5102; benches=7) | Yes | Source | |
| Longcat Flash Cat | - | 89.71% | inferred high-confidence family alias from longcat-flash-chat (score=0.4667; benches=16) | Yes | Source | |
| Kimi K2 (2025-09-05) | 05 Sept 2025 | 89.50% | EM | Yes | Source | |
| Gemini 2.5 Pro Preview (2025-06-05) | 05 Jun 2025 | 89.20% | - | Yes | Source | |
| Qwen 3 VL 235B A22B Instruct | - | 88.80% | - | Yes | - | |
| Qwen 3 VL 32B Thinking | - | 88.70% | - | Yes | - | |
| GPT 4o (2024-05-13) | 13 May 2024 | 88.70% | - | Yes | Source | |
| Gemini 2.5 Pro Preview (2025-05-06) | 06 May 2025 | 88.60% | Lite | Yes | Source | |
| DeepSeek OCR | 20 Oct 2025 | 88.50% | inferred family alias from deepseek-v3 (score=0.3000; benches=20) | Yes | Source | |
| DeepSeek V2 (2024-06-28) | 28 Jun 2024 | 88.50% | inferred family alias from deepseek-v3 (score=0.4159; benches=20) | Yes | Source | |
| DeepSeek V4 | - | 88.50% | inferred high-confidence family alias from deepseek-v3 (score=0.5818; benches=20) | Yes | Source | |
| Gemini 2.5 Flash Preview (2025-04-17) | 17 Apr 2025 | 88.40% | Lite | Yes | Source | |
| Gemini 2.0 Flash | 05 Feb 2025 | 88.40% | Lite | Yes | Source | |
| Gemini 2.5 Flash Preview (2025-05-20) | 20 May 2025 | 88.40% | Lite | Yes | Source | |
| Qwen 3 235B A22B | - | 87.81% | - | Yes | Source | |
| Kimi K2 (2025-07-11) | 11 Jul 2025 | 87.80% | EM | Yes | Source | |
| Qwen 3 VL 30B A3B Thinking | - | 87.60% | - | Yes | - | |
| Grok 2 | 13 Aug 2024 | 87.50% | - | Yes | Source | |
| Kimi K1.5 | 20 Jan 2025 | 87.40% | - | Yes | Source | |
| o3 mini | 30 Jan 2025 | 86.90% | - | Yes | Source | |
| GPT 4 Turbo (2023-11-06) | 06 Nov 2023 | 86.50% | inferred version-family alias from gpt-4-turbo-2024-04-09 | Yes | Source | |
| Qwen 3 VL 32B Instruct | - | 86.40% | - | Yes | - | |
| GPT 4 32K 0613 | - | 86.40% | inferred high-confidence family alias from gpt-4-0613 (score=0.4899; benches=12) | Yes | Source | |
| Grok 2 Mini | 13 Aug 2024 | 86.20% | - | Yes | Source | |
| Llama 2 70B Chat | 20 Jun 2023 | 86% | inferred family alias from llama-3.3-70b-instruct (score=0.3129; benches=9) | Yes | Source | |
| Llama 3.2 90B Vision Instruct | - | 86% | inferred high-confidence family alias from llama-3.2-90b-instruct (score=0.5873; benches=13) | Yes | Source | |
| LearnLM 1.5 Pro Experimental | 19 Nov 2024 | 85.90% | inferred family alias from gemini-1.5-pro (score=0.3700; benches=23) | Yes | Source | |
| Gemini 1.5 Pro 001 | 23 May 2024 | 85.90% | inferred alias from gemini-1.5-pro | Yes | Source | |
| Gemini 1.5 Pro Exp (2024-08-27) | 27 Aug 2024 | 85.90% | inferred alias from gemini-1.5-pro | Yes | Source | |
| Gemini 1.5 Pro Exp (2024-08-01) | 01 Aug 2024 | 85.90% | inferred alias from gemini-1.5-pro | Yes | Source | |
| Gemini Robotics ER 1.5 Preview | 25 Sept 2025 | 85.90% | inferred family alias from gemini-1.5-pro (score=0.3717; benches=23) | Yes | Source | |
| GPT 4o Realtime Preview (2024-10-01) | 01 Oct 2024 | 85.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Search Preview | 11 Mar 2025 | 85.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o (2024-08-06) | 06 Aug 2024 | 85.70% | - | Yes | Source | |
| GPT 4o Realtime Preview (2025-06-03) | 03 Jun 2025 | 85.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Transcribe | 20 Mar 2025 | 85.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Audio (2024-12-17) | 17 Dec 2024 | 85.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Audio (2025-06-03) | 03 Jun 2025 | 85.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Audio (2024-10-01) | 01 Oct 2024 | 85.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| GPT 4o Transcribe Diarize | 15 Oct 2025 | 85.70% | inferred modality/version alias from gpt-4o-2024-08-06 | Yes | Source | |
| Llama 4 Maverick | 05 Apr 2025 | 85.50% | - | Yes | Source | |
| GPT OSS 20b | 05 Aug 2025 | 85.30% | High Reasoning Effort | Yes | Source | |
| GPT OSS Safeguard 20b | 29 Oct 2025 | 85.30% | inferred high-confidence family alias from gpt-oss-20b (score=0.5137; benches=7) | Yes | Source | |
| o1 mini | 12 Sept 2024 | 85.20% | - | Yes | Source | |
| Qwen 3 Guard Gen 8B | - | 85.20% | inferred family alias from qwen3-vl-8b-thinking (score=0.3400; benches=50) | Yes | - | |
| Qwen 3 VL 8B Thinking | - | 85.20% | - | Yes | - | |
| Qwen 3 VL Embedding 8B | - | 85.20% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5232; benches=50) | Yes | - | |
| Qwen 3 8B | - | 85.20% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50) | Yes | - | |
| Qwen 3 Embedding 8B | - | 85.20% | inferred family alias from qwen3-vl-8b-thinking (score=0.3850; benches=50) | Yes | - | |
| Qwen 3 Guard Stream 8B | - | 85.20% | inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50) | Yes | - | |
| Qwen 3 Reranker 8B | - | 85.20% | inferred family alias from qwen3-vl-8b-thinking (score=0.3850; benches=50) | Yes | - | |
| Qwen 3 VL Reranker 8B | - | 85.20% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5275; benches=50) | Yes | - | |
| Qwen 3 VL 30B A3B Instruct | - | 85% | - | Yes | - | |
| Phi 1 | - | 84.80% | inferred family alias from phi-4 (score=0.3100; benches=13) | Yes | Source | |
| Phi 2 | - | 84.80% | inferred family alias from phi-4 (score=0.3100; benches=13) | Yes | Source | |
| Phi 4 | 12 Dec 2024 | 84.80% | - | Yes | Source | |
| Voxtral Mini Transcribe 2 | 04 Feb 2026 | 84% | manual fallback alias from mistral-large-2-2407 | Yes | Source | |
| Mistral Moderation 2 | 17 Mar 2026 | 84% | manual fallback alias from mistral-large-2-2407 | Yes | Source | |
| Mistral OCR 2 | 22 May 2025 | 84% | inferred family alias from mistral-large-2-2407 (score=0.3100; benches=5) | Yes | Source | |
| Mistral Large 2.1 | 18 Nov 2024 | 84% | inferred version-family alias from mistral-large-2-2407 | Yes | Source | |
| Gemini 1.0 Ultra | 06 Dec 2023 | 83.70% | - | No | Source | |
| Qwen 2 Math RM 72B | - | 82.30% | inferred family alias from qwen2-72b-instruct (score=0.3917; benches=17) | Yes | Source | |
| Qwen 2 Math 72B | - | 82.30% | inferred high-confidence family alias from qwen2-72b-instruct (score=0.4667; benches=17) | Yes | Source | |
| GPT 4o Mini Transcribe (2025-03-20) | 20 Mar 2025 | 82% | inferred modality/version alias from gpt-4o-mini-2024-07-18 | Yes | Source | |
| GPT 4o Mini Audio Preview | 17 Dec 2024 | 82% | inferred modality/version alias from gpt-4o-mini-2024-07-18 | Yes | Source | |
| GPT 4o Mini Transcribe (2025-12-15) | 15 Dec 2025 | 82% | inferred modality/version alias from gpt-4o-mini-2024-07-18 | Yes | Source | |
| GPT 4o Mini Realtime Preview | 17 Dec 2024 | 82% | inferred modality/version alias from gpt-4o-mini-2024-07-18 | Yes | Source | |
| GPT 4o Mini TTS (2025-03-20) | 20 Mar 2025 | 82% | inferred modality/version alias from gpt-4o-mini-2024-07-18 | Yes | Source | |
| GPT 4o Mini Search Preview | 11 Mar 2025 | 82% | inferred modality/version alias from gpt-4o-mini-2024-07-18 | Yes | Source | |
| GPT 4o Mini TTS (2025-12-15) | 15 Dec 2025 | 82% | inferred modality/version alias from gpt-4o-mini-2024-07-18 | Yes | Source | |
| Qwen 3 4B SafeRL | - | 81.50% | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | Yes | - | |
| Qwen 3 Guard Gen 4B | - | 81.50% | inferred family alias from qwen3-vl-4b-thinking (score=0.3400; benches=48) | Yes | - | |
| Qwen 3 Embedding 4B | - | 81.50% | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | Yes | - | |
| Qwen 3 Guard Stream 4B | - | 81.50% | inferred family alias from qwen3-vl-4b-thinking (score=0.3371; benches=48) | Yes | - | |
| Qwen 3 4B Instruct 2507 | - | 81.50% | inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48) | Yes | - | |
| Qwen 3 4B Thinking 2507 | - | 81.50% | inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48) | Yes | - | |
| Qwen 3 Reranker 4B | - | 81.50% | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | Yes | - | |
| Qwen 3 VL 4B Thinking | - | 81.50% | - | Yes | - | |
| Grok 1.5 | 28 Mar 2024 | 81.30% | - | Yes | Source | |
| Qwen 3 VL 8B Instruct | - | 80.70% | - | Yes | - | |
| Mistral Small 3.2 | 20 Jun 2025 | 80.50% | - | Yes | Source | |
| DeepSeek V2.5 (2024-12-10) | 10 Dec 2024 | 80.40% | inferred alias from deepseek-v2.5 | Yes | Source | |
| DeepSeek V2.5 (2024-09-05) | 05 Sept 2024 | 80.40% | inferred alias from deepseek-v2.5 | Yes | Source | |
| Llama 3.1 Nemotron 70B Instruct | 01 Oct 2024 | 80.20% | - | Yes | Source | |
| Qwen 14B | - | 79.70% | inferred family alias from qwen-2.5-14b-instruct (score=0.3060; benches=16) | Yes | Source | |
| Llama 4 Scout | 05 Apr 2025 | 79.60% | - | Yes | Source | |
| Gemini 1.5 Flash 001 | 23 May 2024 | 78.90% | inferred alias from gemini-1.5-flash | Yes | Source | |
| Gemini 1.5 Flash Preview | 14 May 2024 | 78.90% | inferred alias from gemini-1.5-flash | Yes | Source | |
| Phi 3.5 MoE instruct | 23 Aug 2024 | 78.90% | - | Yes | Source | |
| Qwen 2.5 Coder 32B Instruct | - | 78.40% | inferred high-confidence family alias from qwen2.5-vl-32b (score=0.4641; benches=28) | Yes | Source | |
| Qwen 2.5 VL 32B Instruct | - | 78.40% | - | Yes | Source | |
| Qwen 2.5 VL 3B Instruct | - | 78.40% | inferred high-confidence family alias from qwen2.5-vl-32b (score=0.4914; benches=28) | Yes | Source | |
| Qwen 3 VL 4B Instruct | - | 77.20% | - | Yes | - | |
| Llama 3.2 11B Vision Instruct | - | 73% | inferred high-confidence family alias from llama-3.2-11b-instruct (score=0.5873; benches=11) | Yes | Source | |
| Llama 3.2 1B Instruct | 25 Sept 2024 | 73% | inferred family alias from llama-3.2-11b-instruct (score=0.4200; benches=11) | Yes | Source | |
| Grok 1 | 03 Nov 2023 | 73% | 5 Shot | Yes | Source | |
| Gemini 1.0 Nano | 06 Dec 2023 | 71.80% | inferred modality/version alias from gemini-1.0-pro | Yes | Source | |
| Gemini 1.0 Pro | 06 Dec 2023 | 71.80% | - | Yes | Source | |
| Gemini 1.0 Pro Vision 001 | 15 Feb 2024 | 71.80% | inferred modality/version alias from gemini-1.0-pro | Yes | Source | |
| Qwen 2 Audio 7B | - | 70.50% | inferred modality/version alias from qwen2-7b-instruct | Yes | Source | |
| Qwen 2 Math 7B | - | 70.50% | inferred high-confidence family alias from qwen2-7b-instruct (score=0.4706; benches=14) | Yes | Source | |
| GPT 3.5 Turbo 16K 0613 | 13 Jun 2023 | 69.80% | inferred family alias from gpt-3.5-turbo-0125 (score=0.4244; benches=8) | No | Source | |
| GPT 3.5 Turbo (2023-11-06) | 28 Sept 2023 | 69.80% | inferred modality/version alias from gpt-3.5-turbo-0125 | No | Source | |
| GPT 3.5 Turbo Instruct | 06 Nov 2023 | 69.80% | inferred high-confidence family alias from gpt-3.5-turbo-0125 (score=0.5539; benches=8) | No | Source | |
| Pixtral 12B | 17 Sept 2024 | 69.20% | inferred version-family alias from pixtral-12b-2409 | Yes | Source | |
| Gemini Diffusion | 20 May 2025 | 69.10% | Pass@1 | Yes | Source | |
| Phi 3 Mini 128K Instruct | - | 69% | inferred family alias from phi-3.5-mini-instruct (score=0.3533; benches=31) | Yes | Source | |
| Phi 3.5 mini instruct | 23 Aug 2024 | 69% | - | Yes | Source | |
| Mistral Nemo 12B | 18 Jul 2024 | 68% | inferred family alias from mistral-nemo-instruct-2407 (score=0.3250; benches=8) | Yes | Source | |
| Phi 4 Mini | 01 Feb 2025 | 67.30% | - | Yes | Source | |
| Grok 0 | - | 65.70% | 5 Shot | Yes | Source | |
| Granite 3.2 8B Instruct | - | 65.54% | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14) | Yes | Source | |
| Granite Guardian 3.0 8B | - | 65.54% | inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14) | Yes | Source | |
| Granite Guardian 3.3 8B | - | 65.54% | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.5071; benches=14) | Yes | Source | |
| Granite 3.3 8B Instruct | 16 Apr 2025 | 65.54% | - | Yes | Source | |
| Granite 3.2 8B Instruct Preview | - | 65.54% | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4687; benches=14) | Yes | Source | |
| Granite 3.1 8B Instruct | - | 65.54% | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14) | Yes | Source | |
| Granite 3.3 2B Instruct | 16 Apr 2025 | 65.54% | inferred family alias from granite-3.3-8b-instruct (score=0.3627; benches=14) | Yes | Source | |
| Granite Guardian 3.1 8B | - | 65.54% | inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14) | Yes | Source | |
| Granite 3.0 8B Instruct | - | 65.54% | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14) | Yes | Source | |
| Granite Speech 3.2 8B | - | 65.54% | inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14) | Yes | Source | |
| Granite Speech 3.3 8B | - | 65.54% | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.5071; benches=14) | Yes | Source | |
| Granite 4.0 Tiny | 02 Oct 2025 | 60.40% | inferred alias from granite-4.0-tiny-preview | Yes | Source | |
| Granite 4.0 Micro | 02 Oct 2025 | 60.40% | inferred high-confidence family alias from granite-4.0-tiny-preview (score=0.4700; benches=12) | Yes | Source | |
| Granite 4.0 Small | 02 Oct 2025 | 60.40% | inferred high-confidence family alias from granite-4.0-tiny-preview (score=0.4700; benches=12) | Yes | Source | |
| Granite 4.0 Tiny Preview | 02 May 2025 | 60.40% | - | Yes | Source | |
| Ernie 4.5 21B A3B | - | 41.90% | inferred version-family alias from ernie-4.5 | Yes | Source | |
| Ernie 4.5 Turbo | - | 41.90% | inferred version-family alias from ernie-4.5 | Yes | Source | |
| Ernie 4.5 21B A3B Thinking | - | 41.90% | inferred version-family alias from ernie-4.5 | Yes | Source | |
| Ernie 4.5 300B A47B | - | 41.90% | inferred version-family alias from ernie-4.5 | Yes | Source | |
| Ernie 4.5 VL 28B A3B | - | 41.90% | inferred version-family alias from ernie-4.5 | Yes | Source | |
| Ernie 4.5 VL 424B A47B | - | 41.90% | inferred version-family alias from ernie-4.5 | Yes | Source |
