AI benchmarks and leaderboards are different things
Learn the difference between a benchmark, an evaluation run and a leaderboard, and use each to answer the right question.
Artificial Analysis
Mathematical reasoning
These are source-reported observations, not independent reruns by this site. Check the task version, evaluation protocol, unit and reporting date before comparing models. Results from different setups are not a universal ranking.
Shown in source observation order, not ranked. Dates retain the precision provided by the source.
| Model | Value | Unit | Protocol | Reported date and basis | Source |
|---|---|---|---|---|---|
| GPT-5.2 (xhigh) | 0.99 | Not recorded | Not recorded | 2025-12-11 · model_announcement | Artificial Analysis |
| GPT-5 Codex (high) | 0.986666666666667 | Not recorded | Not recorded | 2025-09-23 · model_announcement | Artificial Analysis |
| Gemini 3 Flash Preview (Reasoning) | 0.97 | Not recorded | Not recorded | 2025-12-17 · model_announcement | Artificial Analysis |
| DeepSeek V3.2 Speciale | 0.966666666666667 | Not recorded | Not recorded | 2025-12-01 · model_announcement | Artificial Analysis |
| GPT-5.2 (medium) | 0.966666666666667 | Not recorded | Not recorded | 2025-12-11 · model_announcement | Artificial Analysis |
| MiMo-V2-Flash (Reasoning) | 0.963333333333333 | Not recorded | Not recorded | 2025-12-16 · model_announcement | Artificial Analysis |
| GPT-5.1 Codex (high) | 0.956666666666667 | Not recorded | Not recorded | 2025-11-13 · model_announcement | Artificial Analysis |
| Gemini 3 Pro Preview (high) | 0.956666666666667 | Not recorded | Not recorded | 2025-11-18 · model_announcement | Artificial Analysis |
| GLM-4.7 (Reasoning) | 0.95 | Not recorded | Not recorded | 2025-12-22 · model_announcement | Artificial Analysis |
| Kimi K2 Thinking | 0.946666666666667 | Not recorded | Not recorded | 2025-11-06 · model_announcement | Artificial Analysis |
| KAT-Coder-Pro V1 | 0.946666666666667 | Not recorded | Not recorded | 2025-11-11 · model_announcement | Artificial Analysis |
| GPT-5 (high) | 0.943333333333333 | Not recorded | Not recorded | 2025-08-07 · model_announcement | Artificial Analysis |
| Nova 2.0 Lite (high) | 0.943333333333333 | Not recorded | Not recorded | 2025-10-29 · model_announcement | Artificial Analysis |
| GPT-5.1 (high) | 0.94 | Not recorded | Not recorded | 2025-11-13 · model_announcement | Artificial Analysis |
| gpt-oss-120b (high) | 0.934416666666667 | Not recorded | Not recorded | 2025-08-05 · model_announcement | Artificial Analysis |
| Grok 4 | 0.926666666666667 | Not recorded | Not recorded | 2025-07-10 · model_announcement | Artificial Analysis |
| DeepSeek V3.2 (Reasoning) | 0.92 | Not recorded | Not recorded | 2025-12-01 · model_announcement | Artificial Analysis |
| GPT-5 (medium) | 0.916666666666667 | Not recorded | Not recorded | 2025-08-07 · model_announcement | Artificial Analysis |
| GPT-5.1 Codex mini (high) | 0.916666666666667 | Not recorded | Not recorded | 2025-11-13 · model_announcement | Artificial Analysis |
| Claude Opus 4.5 (Reasoning) | 0.913333333333333 | Not recorded | Not recorded | 2025-11-24 · model_announcement | Artificial Analysis |
| NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) | 0.91 | Not recorded | Not recorded | 2025-12-15 · model_announcement | Artificial Analysis |
| Qwen3 235B A22B 2507 (Reasoning) | 0.91 | Not recorded | Not recorded | 2025-07-25 · model_announcement | Artificial Analysis |
| GPT-5 mini (high) | 0.906666666666667 | Not recorded | Not recorded | 2025-08-07 · model_announcement | Artificial Analysis |
| o4-mini (high) | 0.906666666666667 | Not recorded | Not recorded | 2025-04-16 · model_announcement | Artificial Analysis |
| K-EXAONE (Reasoning) | 0.903333333333334 | Not recorded | Not recorded | 2025-12-31 · model_announcement | Artificial Analysis |
| Nova 2.0 Omni (medium) | 0.896666666666667 | Not recorded | Not recorded | 2025-11-26 · model_announcement | Artificial Analysis |
| DeepSeek V3.1 Terminus (Reasoning) | 0.896666666666667 | Not recorded | Not recorded | 2025-09-22 · model_announcement | Artificial Analysis |
| DeepSeek V3.1 (Reasoning) | 0.896666666666667 | Not recorded | Not recorded | 2025-08-21 · model_announcement | Artificial Analysis |
| Grok 4 Fast (Reasoning) | 0.896666666666667 | Not recorded | Not recorded | 2025-09-19 · model_announcement | Artificial Analysis |
| Grok 4.1 Fast (Reasoning) | 0.893333333333333 | Not recorded | Not recorded | 2025-11-19 · model_announcement | Artificial Analysis |
| gpt-oss-20b (high) | 0.893333333333333 | Not recorded | Not recorded | 2025-08-05 · model_announcement | Artificial Analysis |
| Ring-1T | 0.893333333333333 | Not recorded | Not recorded | 2025-10-13 · model_announcement | Artificial Analysis |
| Nova 2.0 Pro Preview (medium) | 0.89 | Not recorded | Not recorded | 2025-11-27 · model_announcement | Artificial Analysis |
| Nova 2.0 Lite (medium) | 0.886666666666667 | Not recorded | Not recorded | 2025-10-29 · model_announcement | Artificial Analysis |
| o3 | 0.883333333333333 | Not recorded | Not recorded | 2025-04-16 · model_announcement | Artificial Analysis |
| Qwen3 VL 235B A22B (Reasoning) | 0.883333333333333 | Not recorded | Not recorded | 2025-09-23 · model_announcement | Artificial Analysis |
| INTELLECT-3 | 0.88 | Not recorded | Not recorded | 2025-11-27 · model_announcement | Artificial Analysis |
| Claude 4.5 Sonnet (Reasoning) | 0.88 | Not recorded | Not recorded | 2025-09-29 · model_announcement | Artificial Analysis |
| Apriel-v1.6-15B-Thinker | 0.88 | Not recorded | Not recorded | 2025-11-25 · model_announcement | Artificial Analysis |
| Gemini 2.5 Pro | 0.876666666666667 | Not recorded | Not recorded | 2025-06-05 · model_announcement | Artificial Analysis |
| DeepSeek V3.2 Exp (Reasoning) | 0.876666666666667 | Not recorded | Not recorded | 2025-09-29 · model_announcement | Artificial Analysis |
| Apriel-v1.5-15B-Thinker | 0.875 | Not recorded | Not recorded | 2025-09-30 · model_announcement | Artificial Analysis |
| Gemini 3 Pro Preview (low) | 0.866666666666667 | Not recorded | Not recorded | 2025-11-18 · model_announcement | Artificial Analysis |
| GLM-4.6 (Reasoning) | 0.86 | Not recorded | Not recorded | 2025-09-30 · model_announcement | Artificial Analysis |
| GLM-4.6V (Reasoning) | 0.853333333333333 | Not recorded | Not recorded | 2025-12-08 · model_announcement | Artificial Analysis |
| ERNIE 5.0 Thinking Preview | 0.85 | Not recorded | Not recorded | 2025-11-13 · model_announcement | Artificial Analysis |
| GPT-5 mini (medium) | 0.85 | Not recorded | Not recorded | 2025-08-07 · model_announcement | Artificial Analysis |
| Seed-OSS-36B-Instruct | 0.846666666666667 | Not recorded | Not recorded | 2025-08-20 · model_announcement | Artificial Analysis |
| Qwen3 VL 32B (Reasoning) | 0.846666666666667 | Not recorded | Not recorded | 2025-10-21 · model_announcement | Artificial Analysis |
| Grok 3 mini Reasoning (high) | 0.846666666666667 | Not recorded | Not recorded | 2025-02-19 · model_announcement | Artificial Analysis |
| Qwen3 Next 80B A3B (Reasoning) | 0.843333333333333 | Not recorded | Not recorded | 2025-09-11 · model_announcement | Artificial Analysis |
| Ring-flash-2.0 | 0.836666666666667 | Not recorded | Not recorded | 2025-09-19 · model_announcement | Artificial Analysis |
| Claude 4.5 Haiku (Reasoning) | 0.836666666666667 | Not recorded | Not recorded | 2025-10-15 · model_announcement | Artificial Analysis |
| GPT-5 nano (high) | 0.836666666666667 | Not recorded | Not recorded | 2025-08-07 · model_announcement | Artificial Analysis |
| GPT-5 (low) | 0.83 | Not recorded | Not recorded | 2025-08-07 · model_announcement | Artificial Analysis |
| MiniMax-M2.1 | 0.826666666666667 | Not recorded | Not recorded | 2025-12-23 · model_announcement | Artificial Analysis |
| Qwen3 4B 2507 (Reasoning) | 0.826666666666667 | Not recorded | Not recorded | 2025-08-06 · model_announcement | Artificial Analysis |
| Qwen3 Max Thinking (Preview) | 0.823333333333333 | Not recorded | Not recorded | 2025-11-03 · model_announcement | Artificial Analysis |
| Qwen3 VL 30B A3B (Reasoning) | 0.823333333333333 | Not recorded | Not recorded | 2025-10-03 · model_announcement | Artificial Analysis |
| Magistral Medium 1.2 | 0.82 | Not recorded | Not recorded | 2025-09-18 · model_announcement | Artificial Analysis |
| Qwen3 235B A22B (Reasoning) | 0.82 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| GLM-4.5-Air | 0.806666666666667 | Not recorded | Not recorded | 2025-07-28 · model_announcement | Artificial Analysis |
| Qwen3 Max | 0.806666666666667 | Not recorded | Not recorded | 2025-09-23 · model_announcement | Artificial Analysis |
| Motif-2-12.7B-Reasoning | 0.803333333333333 | Not recorded | Not recorded | 2025-12-04 · model_announcement | Artificial Analysis |
| Magistral Small 1.2 | 0.803333333333333 | Not recorded | Not recorded | 2025-09-17 · model_announcement | Artificial Analysis |
| Claude 4.1 Opus (Reasoning) | 0.803333333333333 | Not recorded | Not recorded | 2025-08-05 · model_announcement | Artificial Analysis |
| EXAONE 4.0 32B (Reasoning) | 0.8 | Not recorded | Not recorded | 2025-07-15 · model_announcement | Artificial Analysis |
| Falcon-H1R-7B | 0.8 | Not recorded | Not recorded | 2026-01-04 · model_announcement | Artificial Analysis |
| o1-preview | 0.796666666666667 | Not recorded | Not recorded | 2024-09-12 · model_announcement | Artificial Analysis |
| Doubao Seed Code | 0.793333333333333 | Not recorded | Not recorded | 2025-11-11 · model_announcement | Artificial Analysis |
| Mi:dm K 2.5 Pro Preview | 0.786666666666667 | Not recorded | Not recorded | 2025-12-11 · model_announcement | Artificial Analysis |
| GPT-5 nano (medium) | 0.783333333333333 | Not recorded | Not recorded | 2025-08-07 · model_announcement | Artificial Analysis |
| K2-V2 (high) | 0.783333333333333 | Not recorded | Not recorded | 2025-12-05 · model_announcement | Artificial Analysis |
| Gemini 2.5 Flash Preview (Sep '25) (Reasoning) | 0.783333333333333 | Not recorded | Not recorded | 2025-09-25 · model_announcement | Artificial Analysis |
| MiniMax-M2 | 0.783333333333333 | Not recorded | Not recorded | 2025-10-26 · model_announcement | Artificial Analysis |
| Olmo 3.1 32B Think | 0.773333333333333 | Not recorded | Not recorded | 2025-12-12 · model_announcement | Artificial Analysis |
| Mi:dm K 2.5 Pro | 0.766666666666667 | Not recorded | Not recorded | 2025-12-11 · model_announcement | Artificial Analysis |
| Llama Nemotron Super 49B v1.5 (Reasoning) | 0.766666666666667 | Not recorded | Not recorded | 2025-07-25 · model_announcement | Artificial Analysis |
| DeepSeek R1 0528 (May '25) | 0.76 | Not recorded | Not recorded | 2025-05-28 · model_announcement | Artificial Analysis |
| Qwen3 Max (Preview) | 0.75 | Not recorded | Not recorded | 2025-09-05 · model_announcement | Artificial Analysis |
| NVIDIA Nemotron Nano 12B v2 VL (Reasoning) | 0.75 | Not recorded | Not recorded | 2025-10-28 · model_announcement | Artificial Analysis |
| Claude 4 Sonnet (Reasoning) | 0.743333333333333 | Not recorded | Not recorded | 2025-05-22 · model_announcement | Artificial Analysis |
| Qwen3 Omni 30B A3B (Reasoning) | 0.74 | Not recorded | Not recorded | 2025-09-22 · model_announcement | Artificial Analysis |
| GLM-4.5 (Reasoning) | 0.736666666666667 | Not recorded | Not recorded | 2025-07-28 · model_announcement | Artificial Analysis |
| Olmo 3 32B Think | 0.736666666666667 | Not recorded | Not recorded | 2025-11-20 · model_announcement | Artificial Analysis |
| Gemini 2.5 Flash (Reasoning) | 0.733333333333333 | Not recorded | Not recorded | 2025-05-20 · model_announcement | Artificial Analysis |
| Claude 4 Opus (Reasoning) | 0.733333333333333 | Not recorded | Not recorded | 2025-05-22 · model_announcement | Artificial Analysis |
| GLM-4.5V (Reasoning) | 0.73 | Not recorded | Not recorded | 2025-08-11 · model_announcement | Artificial Analysis |
| Qwen3 32B (Reasoning) | 0.73 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| Cogito v2.1 (Reasoning) | 0.726666666666667 | Not recorded | Not recorded | 2025-11-18 · model_announcement | Artificial Analysis |
| Qwen3 30B A3B (Reasoning) | 0.723333333333333 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| Qwen3 VL 30B A3B Instruct | 0.723333333333333 | Not recorded | Not recorded | 2025-10-03 · model_announcement | Artificial Analysis |
| Qwen3 235B A22B 2507 Instruct | 0.716666666666667 | Not recorded | Not recorded | 2025-07-21 · model_announcement | Artificial Analysis |
| Ling-1T | 0.713333333333333 | Not recorded | Not recorded | 2025-10-08 · model_announcement | Artificial Analysis |
| Olmo 3 7B Think | 0.706666666666667 | Not recorded | Not recorded | 2025-11-20 · model_announcement | Artificial Analysis |
| Qwen3 VL 235B A22B Instruct | 0.706666666666667 | Not recorded | Not recorded | 2025-09-23 · model_announcement | Artificial Analysis |
| Hermes 4 - Llama-3.1 405B (Reasoning) | 0.696666666666667 | Not recorded | Not recorded | 2025-08-27 · model_announcement | Artificial Analysis |
| NVIDIA Nemotron Nano 9B V2 (Reasoning) | 0.696666666666667 | Not recorded | Not recorded | 2025-08-18 · model_announcement | Artificial Analysis |
| Hermes 4 - Llama-3.1 70B (Reasoning) | 0.686666666666667 | Not recorded | Not recorded | 2025-08-27 · model_announcement | Artificial Analysis |
| Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) | 0.686666666666667 | Not recorded | Not recorded | 2025-09-08 · model_announcement | Artificial Analysis |
| Qwen3 VL 32B Instruct | 0.683333333333333 | Not recorded | Not recorded | 2025-10-21 · model_announcement | Artificial Analysis |
| DeepSeek R1 (Jan '25) | 0.68 | Not recorded | Not recorded | 2025-01-20 · model_announcement | Artificial Analysis |
| MiMo-V2-Flash (Non-reasoning) | 0.676666666666667 | Not recorded | Not recorded | 2025-12-16 · model_announcement | Artificial Analysis |
| gpt-oss-120b (low) | 0.666666666666667 | Not recorded | Not recorded | 2025-08-05 · model_announcement | Artificial Analysis |
| Qwen3 Next 80B A3B Instruct | 0.663333333333333 | Not recorded | Not recorded | 2025-09-11 · model_announcement | Artificial Analysis |
| Qwen3 30B A3B 2507 Instruct | 0.663333333333333 | Not recorded | Not recorded | 2025-07-29 · model_announcement | Artificial Analysis |
| Ling-flash-2.0 | 0.653333333333333 | Not recorded | Not recorded | 2025-09-17 · model_announcement | Artificial Analysis |
| K2-V2 (medium) | 0.646666666666667 | Not recorded | Not recorded | 2025-12-05 · model_announcement | Artificial Analysis |
| DeepSeek R1 0528 Qwen3 8B | 0.636666666666667 | Not recorded | Not recorded | 2025-05-29 · model_announcement | Artificial Analysis |
| Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) | 0.636666666666667 | Not recorded | Not recorded | 2025-04-07 · model_announcement | Artificial Analysis |
| Nova 2.0 Pro Preview (low) | 0.633333333333333 | Not recorded | Not recorded | 2025-11-27 · model_announcement | Artificial Analysis |
| DeepSeek R1 Distill Qwen 32B | 0.63 | Not recorded | Not recorded | 2025-01-20 · model_announcement | Artificial Analysis |
| Claude Opus 4.5 (Non-reasoning) | 0.626666666666667 | Not recorded | Not recorded | 2025-11-24 · model_announcement | Artificial Analysis |
| gpt-oss-20b (low) | 0.623333333333333 | Not recorded | Not recorded | 2025-08-05 · model_announcement | Artificial Analysis |
| NVIDIA Nemotron Nano 9B V2 (Non-reasoning) | 0.623333333333333 | Not recorded | Not recorded | 2025-08-18 · model_announcement | Artificial Analysis |
| Solar Pro 2 (Reasoning) | 0.613333333333333 | Not recorded | Not recorded | 2025-07-09 · model_announcement | Artificial Analysis |
| MiniMax M1 80k | 0.61 | Not recorded | Not recorded | 2025-06-17 · model_announcement | Artificial Analysis |
| Gemini 2.5 Flash (Non-reasoning) | 0.603333333333333 | Not recorded | Not recorded | 2025-05-20 · model_announcement | Artificial Analysis |
| HyperCLOVA X SEED Think (32B) | 0.59 | Not recorded | Not recorded | 2025-12-26 · model_announcement | Artificial Analysis |
| DeepSeek V3.2 (Non-reasoning) | 0.59 | Not recorded | Not recorded | 2025-12-01 · model_announcement | Artificial Analysis |
| Grok 3 | 0.58 | Not recorded | Not recorded | 2025-02-19 · model_announcement | Artificial Analysis |
| Qwen3 14B (Non-reasoning) | 0.58 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| DeepSeek V3.2 Exp (Non-reasoning) | 0.576666666666667 | Not recorded | Not recorded | 2025-09-29 · model_announcement | Artificial Analysis |
| Kimi K2 0905 | 0.573333333333333 | Not recorded | Not recorded | 2025-09-05 · model_announcement | Artificial Analysis |
| Kimi K2 | 0.57 | Not recorded | Not recorded | 2025-07-11 · model_announcement | Artificial Analysis |
| Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning) | 0.566666666666667 | Not recorded | Not recorded | 2025-09-25 · model_announcement | Artificial Analysis |
| Claude 3.7 Sonnet (Reasoning) | 0.563333333333333 | Not recorded | Not recorded | 2025-02-24 · model_announcement | Artificial Analysis |
| Qwen3 30B A3B 2507 (Reasoning) | 0.563333333333333 | Not recorded | Not recorded | 2025-07-30 · model_announcement | Artificial Analysis |
| Nova 2.0 Omni (low) | 0.56 | Not recorded | Not recorded | 2025-11-26 · model_announcement | Artificial Analysis |
| Qwen3 14B (Reasoning) | 0.556666666666667 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| Gemini 3 Flash Preview (Non-reasoning) | 0.556666666666667 | Not recorded | Not recorded | 2025-12-17 · model_announcement | Artificial Analysis |
| DeepSeek R1 Distill Qwen 14B | 0.556666666666667 | Not recorded | Not recorded | 2025-01-20 · model_announcement | Artificial Analysis |
| Llama 3.3 Nemotron Super 49B v1 (Reasoning) | 0.546666666666667 | Not recorded | Not recorded | 2025-03-18 · model_announcement | Artificial Analysis |
| DeepSeek R1 Distill Llama 70B | 0.536666666666667 | Not recorded | Not recorded | 2025-01-20 · model_announcement | Artificial Analysis |
| DeepSeek V3.1 Terminus (Non-reasoning) | 0.536666666666667 | Not recorded | Not recorded | 2025-09-22 · model_announcement | Artificial Analysis |
| Gemini 2.5 Flash-Lite (Reasoning) | 0.533333333333333 | Not recorded | Not recorded | 2025-06-17 · model_announcement | Artificial Analysis |
| Qwen3 Omni 30B A3B Instruct | 0.523333333333333 | Not recorded | Not recorded | 2025-09-22 · model_announcement | Artificial Analysis |
| Qwen3 4B 2507 Instruct | 0.523333333333333 | Not recorded | Not recorded | 2025-08-06 · model_announcement | Artificial Analysis |
| GPT-5.2 (Non-reasoning) | 0.51 | Not recorded | Not recorded | 2025-12-11 · model_announcement | Artificial Analysis |
| Exaone 4.0 1.2B (Reasoning) | 0.503333333333333 | Not recorded | Not recorded | 2025-07-15 · model_announcement | Artificial Analysis |
| Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning) | 0.5 | Not recorded | Not recorded | 2025-05-20 · model_announcement | Artificial Analysis |
| DeepSeek V3.1 (Non-reasoning) | 0.496666666666667 | Not recorded | Not recorded | 2025-08-21 · model_announcement | Artificial Analysis |
| Ling-mini-2.0 | 0.493333333333333 | Not recorded | Not recorded | 2025-09-09 · model_announcement | Artificial Analysis |
| GPT-5 (ChatGPT) | 0.483333333333333 | Not recorded | Not recorded | 2025-08-07 · model_announcement | Artificial Analysis |
| GLM-4.7 (Non-reasoning) | 0.48 | Not recorded | Not recorded | 2025-12-22 · model_announcement | Artificial Analysis |
| Nova 2.0 Lite (low) | 0.466666666666667 | Not recorded | Not recorded | 2025-10-29 · model_announcement | Artificial Analysis |
| Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) | 0.466666666666667 | Not recorded | Not recorded | 2025-09-25 · model_announcement | Artificial Analysis |
| GPT-5 mini (minimal) | 0.466666666666667 | Not recorded | Not recorded | 2025-08-07 · model_announcement | Artificial Analysis |
| GPT-4.1 mini | 0.463333333333333 | Not recorded | Not recorded | 2025-04-14 · model_announcement | Artificial Analysis |
| GLM-4.6 (Non-reasoning) | 0.443333333333333 | Not recorded | Not recorded | 2025-09-30 · model_announcement | Artificial Analysis |
| K-EXAONE (Non-reasoning) | 0.44 | Not recorded | Not recorded | 2025-12-31 · model_announcement | Artificial Analysis |
| Grok Code Fast 1 | 0.433333333333333 | Not recorded | Not recorded | 2025-08-28 · model_announcement | Artificial Analysis |
| DeepSeek R1 Distill Llama 8B | 0.413333333333333 | Not recorded | Not recorded | 2025-01-20 · model_announcement | Artificial Analysis |
| Magistral Small 1 | 0.413333333333333 | Not recorded | Not recorded | 2025-06-10 · model_announcement | Artificial Analysis |
| Grok 4 Fast (Non-reasoning) | 0.413333333333333 | Not recorded | Not recorded | 2025-09-19 · model_announcement | Artificial Analysis |
| Olmo 3 7B Instruct | 0.413333333333333 | Not recorded | Not recorded | 2025-11-20 · model_announcement | Artificial Analysis |
| ERNIE 4.5 300B A47B | 0.413333333333333 | Not recorded | Not recorded | 2025-06-30 · model_announcement | Artificial Analysis |
| DeepSeek V3 0324 | 0.41 | Not recorded | Not recorded | 2025-03-25 · model_announcement | Artificial Analysis |
| Magistral Medium 1 | 0.403333333333333 | Not recorded | Not recorded | 2025-06-10 · model_announcement | Artificial Analysis |
| Qwen3 Coder 480B A35B Instruct | 0.393333333333333 | Not recorded | Not recorded | 2025-07-22 · model_announcement | Artificial Analysis |
| EXAONE 4.0 32B (Non-reasoning) | 0.393333333333333 | Not recorded | Not recorded | 2025-07-15 · model_announcement | Artificial Analysis |
| Claude 4.5 Haiku (Non-reasoning) | 0.39 | Not recorded | Not recorded | 2025-10-15 · model_announcement | Artificial Analysis |
| Qwen3 1.7B (Reasoning) | 0.386666666666667 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| Mistral Medium 3.1 | 0.383333333333333 | Not recorded | Not recorded | 2025-08-12 · model_announcement | Artificial Analysis |
| Mistral Large 3 | 0.38 | Not recorded | Not recorded | 2025-12-02 · model_announcement | Artificial Analysis |
| Claude 4 Sonnet (Non-reasoning) | 0.38 | Not recorded | Not recorded | 2025-05-22 · model_announcement | Artificial Analysis |
| GPT-5.1 (Non-reasoning) | 0.38 | Not recorded | Not recorded | 2025-11-13 · model_announcement | Artificial Analysis |
| Nova 2.0 Omni (Non-reasoning) | 0.37 | Not recorded | Not recorded | 2025-11-26 · model_announcement | Artificial Analysis |
| Claude 4.5 Sonnet (Non-reasoning) | 0.37 | Not recorded | Not recorded | 2025-09-29 · model_announcement | Artificial Analysis |
| Qwen3 VL 4B Instruct | 0.37 | Not recorded | Not recorded | 2025-10-14 · model_announcement | Artificial Analysis |
| Devstral 2 | 0.366666666666667 | Not recorded | Not recorded | 2025-12-09 · model_announcement | Artificial Analysis |
| Claude 4 Opus (Non-reasoning) | 0.363333333333333 | Not recorded | Not recorded | 2025-05-22 · model_announcement | Artificial Analysis |
| Kimi Linear 48B A3B Instruct | 0.363333333333333 | Not recorded | Not recorded | 2025-10-30 · model_announcement | Artificial Analysis |
| Gemini 2.5 Flash-Lite (Non-reasoning) | 0.353333333333333 | Not recorded | Not recorded | 2025-06-17 · model_announcement | Artificial Analysis |
| K2-V2 (low) | 0.353333333333333 | Not recorded | Not recorded | 2025-12-05 · model_announcement | Artificial Analysis |
| GPT-4.1 | 0.346666666666667 | Not recorded | Not recorded | 2025-04-14 · model_announcement | Artificial Analysis |
| Grok 4.1 Fast (Non-reasoning) | 0.343333333333333 | Not recorded | Not recorded | 2025-11-19 · model_announcement | Artificial Analysis |
| Devstral Small 2 | 0.343333333333333 | Not recorded | Not recorded | 2025-12-09 · model_announcement | Artificial Analysis |
| Nova 2.0 Lite (Non-reasoning) | 0.336666666666667 | Not recorded | Not recorded | 2025-10-29 · model_announcement | Artificial Analysis |
| Reka Flash 3 | 0.336666666666667 | Not recorded | Not recorded | 2025-03-10 · model_announcement | Artificial Analysis |
| Ministral 3 8B | 0.316666666666667 | Not recorded | Not recorded | 2025-12-02 · model_announcement | Artificial Analysis |
| GPT-5 (minimal) | 0.316666666666667 | Not recorded | Not recorded | 2025-08-07 · model_announcement | Artificial Analysis |
| Nova 2.0 Pro Preview (Non-reasoning) | 0.306666666666667 | Not recorded | Not recorded | 2025-11-27 · model_announcement | Artificial Analysis |
| Qwen3 VL 8B (Reasoning) | 0.306666666666667 | Not recorded | Not recorded | 2025-10-14 · model_announcement | Artificial Analysis |
| Mistral Medium 3 | 0.303333333333333 | Not recorded | Not recorded | 2025-05-07 · model_announcement | Artificial Analysis |
| Solar Pro 2 (Non-reasoning) | 0.3 | Not recorded | Not recorded | 2025-07-09 · model_announcement | Artificial Analysis |
| Ministral 3 14B | 0.3 | Not recorded | Not recorded | 2025-12-02 · model_announcement | Artificial Analysis |
| Devstral Small (Jul '25) | 0.293333333333333 | Not recorded | Not recorded | 2025-07-10 · model_announcement | Artificial Analysis |
| QwQ 32B | 0.29 | Not recorded | Not recorded | 2025-03-05 · model_announcement | Artificial Analysis |
| Qwen3 Coder 30B A3B Instruct | 0.29 | Not recorded | Not recorded | 2025-07-31 · model_announcement | Artificial Analysis |
| GPT-5 nano (minimal) | 0.273333333333333 | Not recorded | Not recorded | 2025-08-07 · model_announcement | Artificial Analysis |
| Qwen3 VL 8B Instruct | 0.273333333333333 | Not recorded | Not recorded | 2025-10-14 · model_announcement | Artificial Analysis |
| Mistral Small 3.2 | 0.27 | Not recorded | Not recorded | 2025-06-20 · model_announcement | Artificial Analysis |
| NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning) | 0.266666666666667 | Not recorded | Not recorded | 2025-10-28 · model_announcement | Artificial Analysis |
| GLM-4.6V (Non-reasoning) | 0.263333333333333 | Not recorded | Not recorded | 2025-12-08 · model_announcement | Artificial Analysis |
| DeepSeek V3 (Dec '24) | 0.26 | Not recorded | Not recorded | 2024-12-26 · model_announcement | Artificial Analysis |
| GPT-4o (March 2025, chatgpt-4o-latest) | 0.256666666666667 | Not recorded | Not recorded | 2025-03-27 · model_announcement | Artificial Analysis |
| Qwen3 VL 4B (Reasoning) | 0.256666666666667 | Not recorded | Not recorded | 2025-10-14 · model_announcement | Artificial Analysis |
| LFM2 8B A1B | 0.253333333333333 | Not recorded | Not recorded | 2025-10-07 · model_announcement | Artificial Analysis |
| Qwen3 8B (Non-reasoning) | 0.243333333333333 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| GPT-4.1 nano | 0.24 | Not recorded | Not recorded | 2025-04-14 · model_announcement | Artificial Analysis |
| Exaone 4.0 1.2B (Non-reasoning) | 0.24 | Not recorded | Not recorded | 2025-07-15 · model_announcement | Artificial Analysis |
| Qwen3 235B A22B (Non-reasoning) | 0.236666666666667 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| Qwen3 4B (Reasoning) | 0.223333333333333 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| DeepSeek R1 Distill Qwen 1.5B | 0.22 | Not recorded | Not recorded | 2025-01-20 · model_announcement | Artificial Analysis |
| Ministral 3 3B | 0.22 | Not recorded | Not recorded | 2025-12-02 · model_announcement | Artificial Analysis |
| Gemini 2.0 Flash (Feb '25) | 0.216666666666667 | Not recorded | Not recorded | 2025-02-05 · model_announcement | Artificial Analysis |
| Qwen3 30B A3B (Non-reasoning) | 0.216666666666667 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| Claude 3.7 Sonnet (Non-reasoning) | 0.21 | Not recorded | Not recorded | 2025-02-24 · model_announcement | Artificial Analysis |
| Gemma 3 27B Instruct | 0.206666666666667 | Not recorded | Not recorded | 2025-03-12 · model_announcement | Artificial Analysis |
| Qwen3 32B (Non-reasoning) | 0.196666666666667 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| Llama 4 Maverick | 0.193333333333333 | Not recorded | Not recorded | 2025-04-05 · model_announcement | Artificial Analysis |
| Qwen3 8B (Reasoning) | 0.19 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| Gemma 3 12B Instruct | 0.183333333333333 | Not recorded | Not recorded | 2025-03-12 · model_announcement | Artificial Analysis |
| Qwen3 0.6B (Reasoning) | 0.18 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| Phi-4 | 0.18 | Not recorded | Not recorded | 2024-12-12 · model_announcement | Artificial Analysis |
| Nova Premier | 0.173333333333333 | Not recorded | Not recorded | 2025-04-30 · model_announcement | Artificial Analysis |
| GLM-4.5V (Non-reasoning) | 0.153333333333333 | Not recorded | Not recorded | 2025-08-11 · model_announcement | Artificial Analysis |
| Hermes 4 - Llama-3.1 405B (Non-reasoning) | 0.153333333333333 | Not recorded | Not recorded | 2025-08-27 · model_announcement | Artificial Analysis |
| GPT-4o mini | 0.146666666666667 | Not recorded | Not recorded | 2024-07-18 · model_announcement | Artificial Analysis |
| Gemma 3n E4B Instruct | 0.143333333333333 | Not recorded | Not recorded | 2025-06-26 · model_announcement | Artificial Analysis |
| Mistral Large 2 (Nov '24) | 0.14 | Not recorded | Not recorded | 2024-11-18 · model_announcement | Artificial Analysis |
| Llama 4 Scout | 0.14 | Not recorded | Not recorded | 2025-04-05 · model_announcement | Artificial Analysis |
| Qwen2.5 Instruct 72B | 0.14 | Not recorded | Not recorded | 2024-09-19 · model_announcement | Artificial Analysis |
| MiniMax M1 40k | 0.136666666666667 | Not recorded | Not recorded | 2025-06-17 · model_announcement | Artificial Analysis |
| Granite 4.0 H Small | 0.136666666666667 | Not recorded | Not recorded | 2025-09-22 · model_announcement | Artificial Analysis |
| NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) | 0.133333333333333 | Not recorded | Not recorded | 2025-12-15 · model_announcement | Artificial Analysis |
| Command A | 0.13 | Not recorded | Not recorded | 2025-03-13 · model_announcement | Artificial Analysis |
| Gemma 3 4B Instruct | 0.126666666666667 | Not recorded | Not recorded | 2025-03-12 · model_announcement | Artificial Analysis |
| Hermes 4 - Llama-3.1 70B (Non-reasoning) | 0.113333333333333 | Not recorded | Not recorded | 2025-08-27 · model_announcement | Artificial Analysis |
| Llama 3.1 Nemotron Instruct 70B | 0.11 | Not recorded | Not recorded | 2024-10-15 · model_announcement | Artificial Analysis |
| Jamba Reasoning 3B | 0.106666666666667 | Not recorded | Not recorded | 2025-10-08 · model_announcement | Artificial Analysis |
| Qwen3 0.6B (Non-reasoning) | 0.103333333333333 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| Gemma 3n E2B Instruct | 0.103333333333333 | Not recorded | Not recorded | 2025-06-26 · model_announcement | Artificial Analysis |
| LFM2 2.6B | 0.0833333333333333 | Not recorded | Not recorded | 2025-09-23 · model_announcement | Artificial Analysis |
| Llama Nemotron Super 49B v1.5 (Non-reasoning) | 0.08 | Not recorded | Not recorded | 2025-07-25 · model_announcement | Artificial Analysis |
| Llama 3.3 Instruct 70B | 0.0766666666666667 | Not recorded | Not recorded | 2024-12-06 · model_announcement | Artificial Analysis |
| Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) | 0.0766666666666667 | Not recorded | Not recorded | 2025-03-18 · model_announcement | Artificial Analysis |
| Qwen3 1.7B (Non-reasoning) | 0.0733333333333333 | Not recorded | Not recorded | 2025-04-28 · model_announcement | Artificial Analysis |
| Nova Pro | 0.07 | Not recorded | Not recorded | 2024-12-03 · model_announcement | Artificial Analysis |
| Nova Lite | 0.07 | Not recorded | Not recorded | 2024-12-03 · model_announcement | Artificial Analysis |
| Granite 3.3 8B (Non-reasoning) | 0.0666666666666667 | Not recorded | Not recorded | 2025-04-16 · model_announcement | Artificial Analysis |
| Phi-4 Mini Instruct | 0.0666666666666667 | Not recorded | Not recorded | 2024-02-26 · model_announcement | Artificial Analysis |
| Granite 4.0 1B | 0.0633333333333333 | Not recorded | Not recorded | 2025-10-28 · model_announcement | Artificial Analysis |
| Granite 4.0 H 1B | 0.0633333333333333 | Not recorded | Not recorded | 2025-10-28 · model_announcement | Artificial Analysis |
| Granite 4.0 Micro | 0.06 | Not recorded | Not recorded | 2025-09-22 · model_announcement | Artificial Analysis |
| Nova Micro | 0.06 | Not recorded | Not recorded | 2024-12-03 · model_announcement | Artificial Analysis |
| GPT-4o (Nov '24) | 0.06 | Not recorded | Not recorded | 2024-11-20 · model_announcement | Artificial Analysis |
| Devstral Medium | 0.0466666666666667 | Not recorded | Not recorded | 2025-07-10 · model_announcement | Artificial Analysis |
| Llama 3.1 Instruct 8B | 0.0433333333333333 | Not recorded | Not recorded | 2024-07-23 · model_announcement | Artificial Analysis |
| Mistral Small 3 | 0.0433333333333333 | Not recorded | Not recorded | 2025-01-30 · model_announcement | Artificial Analysis |
| Llama 3.1 Instruct 70B | 0.04 | Not recorded | Not recorded | 2024-07-23 · model_announcement | Artificial Analysis |
| Mistral Small 3.1 | 0.0366666666666667 | Not recorded | Not recorded | 2025-03-17 · model_announcement | Artificial Analysis |
| LFM2 1.2B | 0.0333333333333333 | Not recorded | Not recorded | 2025-07-10 · model_announcement | Artificial Analysis |
| OLMo 2 32B | 0.0333333333333333 | Not recorded | Not recorded | 2025-03-13 · model_announcement | Artificial Analysis |
| Llama 3.2 Instruct 3B | 0.0333333333333333 | Not recorded | Not recorded | 2024-09-25 · model_announcement | Artificial Analysis |
| Gemma 3 1B Instruct | 0.0333333333333333 | Not recorded | Not recorded | 2025-03-13 · model_announcement | Artificial Analysis |
| Llama 3.1 Instruct 405B | 0.03 | Not recorded | Not recorded | 2024-07-23 · model_announcement | Artificial Analysis |
| Gemma 3 270M | 0.0233333333333333 | Not recorded | Not recorded | 2025-08-14 · model_announcement | Artificial Analysis |
| Jamba 1.7 Large | 0.0233333333333333 | Not recorded | Not recorded | 2025-07-07 · model_announcement | Artificial Analysis |
| Pixtral Large | 0.0233333333333333 | Not recorded | Not recorded | 2024-11-18 · model_announcement | Artificial Analysis |
| Llama 3.2 Instruct 11B (Vision) | 0.0166666666666667 | Not recorded | Not recorded | 2024-09-25 · model_announcement | Artificial Analysis |
| Granite 4.0 H 350M | 0.0133333333333333 | Not recorded | Not recorded | 2025-10-28 · model_announcement | Artificial Analysis |
| OLMo 2 7B | 0.00666666666666667 | Not recorded | Not recorded | 2024-11-26 · model_announcement | Artificial Analysis |
| Jamba 1.7 Mini | 0.00333333333333333 | Not recorded | Not recorded | 2025-07-07 · model_announcement | Artificial Analysis |
| Phi-3 Mini Instruct 3.8B | 0.00333333333333333 | Not recorded | Not recorded | 2024-04-23 · model_announcement | Artificial Analysis |
| Mistral Large 2 (Jul '24) | 0 | Not recorded | Not recorded | 2024-07-24 · model_announcement | Artificial Analysis |
| Granite 4.0 350M | 0 | Not recorded | Not recorded | 2025-10-28 · model_announcement | Artificial Analysis |
| Llama 3.2 Instruct 1B | 0 | Not recorded | Not recorded | 2024-09-25 · model_announcement | Artificial Analysis |
| Molmo 7B-D | 0 | Not recorded | Not recorded | 2024-09-25 · model_announcement | Artificial Analysis |