External benchmark
GPQA diamond
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
Data observatory
General benchmarks help identify candidate models. Our operational work asks whether those capabilities survive the instructions, sources, handoffs and review requirements of a particular task.
Source snapshot retrieved 2026-10-16T15:15:44.060Z. These are source-reported results, not MEKreflect reruns. Coverage is limited to the records available in this release. Missing scores are unknown, never zero.
Sources, reuse and refresh status| Model | Organisation | Measure | Value | Source |
|---|---|---|---|---|
| Claude Opus 5.5 | Anthropic | Epoch Capabilities Index | 166.82 | Source |
| GPT-6 Astra | OpenAI | Epoch Capabilities Index | 166.41 | Source |
| GPT-6.1 Sol | OpenAI | Epoch Capabilities Index | 166.32 | Source |
| Claude Fable 5.1 | Anthropic | Epoch Capabilities Index | 164.65 | Source |
| Gemini 4 Argon | Google DeepMind | Epoch Capabilities Index | 164.56 | Source |
| Claude Sonnet 5.5 | Anthropic | Epoch Capabilities Index | 164.45 | Source |
| Claude Opus 5 | Anthropic | Epoch Capabilities Index | 162.6 | Source |
| GPT-6 Sol | OpenAI | Epoch Capabilities Index | 162.55 | Source |
| Claude Fable 5 | Anthropic | Epoch Capabilities Index | 162.13 | Source |
| GPT-5.5 Pro | OpenAI | Epoch Capabilities Index | 161.97 | Source |
| GPT-5.6 Sol | OpenAI | Epoch Capabilities Index | 161.73 | Source |
| GPT-5.6 Terra | OpenAI | Epoch Capabilities Index | 159.43 | Source |
| GPT-5.4 Pro | OpenAI | Epoch Capabilities Index | 159 | Source |
| Claude Haiku 5.5 | Anthropic | Epoch Capabilities Index | 158.98 | Source |
| GPT-5.5 | OpenAI | Epoch Capabilities Index | 158.95 | Source |
| Claude Opus 4.8 | Anthropic | Epoch Capabilities Index | 158.26 | Source |
| Kimi K3 | Moonshot | Epoch Capabilities Index | 157.42 | Source |
| Gemini 3.7 Flash | Google DeepMind | Epoch Capabilities Index | 157.22 | Source |
| Gemini 3.8 Flash | Google DeepMind | Epoch Capabilities Index | 157.2 | Source |
| Qwen 3.8 Max | Alibaba | Epoch Capabilities Index | 156.88 | Source |
| GPT-5.3 Codex | OpenAI | Epoch Capabilities Index | 156.76 | Source |
| Muse Spark 1.3 | Meta AI | Epoch Capabilities Index | 156.76 | Source |
| GPT-5.4 | OpenAI | Epoch Capabilities Index | 156.72 | Source |
| Claude Sonnet 5 | Anthropic | Epoch Capabilities Index | 156.36 | Source |
| GPT-5.6 Luna | OpenAI | Epoch Capabilities Index | 156.33 | Source |
| Claude Opus 4.7 | Anthropic | Epoch Capabilities Index | 156.32 | Source |
| GPT-6 Luna | OpenAI | Epoch Capabilities Index | 156.31 | Source |
| Grok 4.6 | xAI | Epoch Capabilities Index | 156.31 | Source |
| GPT-5.2 Pro | OpenAI | Epoch Capabilities Index | 156.03 | Source |
| GLM-5.3 | Z.ai (Zhipu AI) | Epoch Capabilities Index | 155.91 | Source |
| DeepSeek V4.1 Flash | DeepSeek | Epoch Capabilities Index | 155.57 | Source |
| DeepSeek V4 Pro 0813 | DeepSeek | Epoch Capabilities Index | 155.29 | Source |
| Claude Opus 4.6 | Anthropic | Epoch Capabilities Index | 155.27 | Source |
| Qwen3.8 Max (0902) | Alibaba | Epoch Capabilities Index | 154.91 | Source |
| Muse Spark 1.2 | Meta AI | Epoch Capabilities Index | 154.9 | Source |
| Gemini 3.1 Pro | Google DeepMind | Epoch Capabilities Index | 154.67 | Source |
| DeepSeek V4 Flash 0731 | DeepSeek | Epoch Capabilities Index | 154.49 | Source |
| Gemini 3.5 Flash | Google DeepMind | Epoch Capabilities Index | 154.38 | Source |
| Grok 4.7 | xAI | Epoch Capabilities Index | 154.37 | Source |
| Gemini 3.6 Flash | Google DeepMind | Epoch Capabilities Index | 154.24 | Source |
| Muse Spark 1.1 | Meta AI | Epoch Capabilities Index | 154 | Source |
| Grok 4.5 | xAI | Epoch Capabilities Index | 153.93 | Source |
| Qwen3.7-Max | Alibaba | Epoch Capabilities Index | 153.83 | Source |
| GPT-5.2 | OpenAI | Epoch Capabilities Index | 153.46 | Source |
| GLM-5.3-Flash | Z.ai (Zhipu AI) | Epoch Capabilities Index | 153.23 | Source |
| Gemini 3 Pro | Google DeepMind | Epoch Capabilities Index | 152.9 | Source |
| Claude Sonnet 4.6 | Anthropic | Epoch Capabilities Index | 152.39 | Source |
| Muse Spark | Meta AI | Epoch Capabilities Index | 152.13 | Source |
| Grok 4.20 | xAI | Epoch Capabilities Index | 152.08 | Source |
| GLM-5.2 | Z.ai (Zhipu AI) | Epoch Capabilities Index | 151.86 | Source |
| Gemini 3 Flash | Google DeepMind | Epoch Capabilities Index | 151.8 | Source |
| Qwen 3.8 27B | Alibaba | Epoch Capabilities Index | 151.55 | Source |
| Kimi K2.6 | Moonshot | Epoch Capabilities Index | 150.92 | Source |
| GPT-5 Pro | OpenAI | Epoch Capabilities Index | 150.74 | Source |
| Inkling-Small | Thinking Machines | Epoch Capabilities Index | 150.18 | Source |
| Claude Opus 4.5 | Anthropic | Epoch Capabilities Index | 150.17 | Source |
| GPT-5 | OpenAI | Epoch Capabilities Index | 150 | Source |
| GLM-5.1 | Z.ai (Zhipu AI) | Epoch Capabilities Index | 149.99 | Source |
| Kimi K2.7 Code | Moonshot | Epoch Capabilities Index | 149.87 | Source |
| GPT-5.1 | OpenAI | Epoch Capabilities Index | 149.76 | Source |
| Qwen 3.6 Max (Preview) | Alibaba | Epoch Capabilities Index | 149.24 | Source |
| DeepSeek-V4-Pro | DeepSeek | Epoch Capabilities Index | 149.19 | Source |
| Grok 4.3 Beta | xAI | Epoch Capabilities Index | 148.95 | Source |
| GPT-5.4 Mini | OpenAI | Epoch Capabilities Index | 148.77 | Source |
| Inkling | Thinking Machines | Epoch Capabilities Index | 148.59 | Source |
| Kimi K2.5 | Moonshot | Epoch Capabilities Index | 147.98 | Source |
| Qwen 3.6 Plus | Alibaba | Epoch Capabilities Index | 147.49 | Source |
| o3-pro | OpenAI | Epoch Capabilities Index | 147.37 | Source |
| Qwen3.7-Plus | Alibaba | Epoch Capabilities Index | 147.3 | Source |
| MiniMax-M3 | MiniMax | Epoch Capabilities Index | 147.13 | Source |
| Mistral Large 4 | Mistral AI | Epoch Capabilities Index | 146.9 | Source |
| o3 | OpenAI | Epoch Capabilities Index | 146.87 | Source |
| Claude Sonnet 4.5 | Anthropic | Epoch Capabilities Index | 146.83 | Source |
| MiniMax-M2.5 | MiniMax | Epoch Capabilities Index | 146.67 | Source |
| Qwen3.5 397B-A17B | Alibaba | Epoch Capabilities Index | 146.66 | Source |
| Qwen3.6 27B | Alibaba | Epoch Capabilities Index | 146.48 | Source |
| Grok 4 | xAI | Epoch Capabilities Index | 146.43 | Source |
| Qwen 3.5 Plus (hosted 397B-A17B) | Alibaba | Epoch Capabilities Index | 146.32 | Source |
| GPT-5.4 Nano | OpenAI | Epoch Capabilities Index | 146.12 | Source |
| DeepSeek-V3.2 | DeepSeek | Epoch Capabilities Index | 146.11 | Source |
| DeepSeek-V4-Flash | DeepSeek | Epoch Capabilities Index | 146.05 | Source |
| Kimi K2 Thinking | Moonshot | Epoch Capabilities Index | 145.96 | Source |
| MiniMax-M2.7 | MiniMax | Epoch Capabilities Index | 145.88 | Source |
| GLM-5 | Z.ai (Zhipu AI) | Epoch Capabilities Index | 145.78 | Source |
| Nemotron 3 Ultra | Nvidia | Epoch Capabilities Index | 145.77 | Source |
| o4-mini | OpenAI | Epoch Capabilities Index | 145.6 | Source |
| GPT-5 mini | OpenAI | Epoch Capabilities Index | 145.43 | Source |
| Gemini 2.5 Pro (Jun 2025) | Google DeepMind | Epoch Capabilities Index | 145.31 | Source |
| Gemini 3.5 Flash-Lite | Google DeepMind | Epoch Capabilities Index | 145.15 | Source |
| DeepSeek-V3.2-Exp | DeepSeek | Epoch Capabilities Index | 144.97 | Source |
| Qwen3.7 Flash | Alibaba | Epoch Capabilities Index | 144.64 | Source |
| Gemini 3.1 Flash-Lite | Epoch Capabilities Index | 144.47 | Source | |
| Grok 4 Fast | xAI | Epoch Capabilities Index | 144.19 | Source |
| Gemini 2.5 Pro (Mar 2025) | Google DeepMind | Epoch Capabilities Index | 144.12 | Source |
| Qwen 3.6 35B-A3B | Alibaba | Epoch Capabilities Index | 144.11 | Source |
| Claude Opus 4.1 | Anthropic | Epoch Capabilities Index | 144.09 | Source |
| Qwen3-235B-A22B-Thinking (Jul 2025) | Alibaba | Epoch Capabilities Index | 143.81 | Source |
| GLM-4.7 | Z.ai (Zhipu AI) | Epoch Capabilities Index | 143.79 | Source |
| Qwen 3.5 Flash (hosted 35B-A3B) | Alibaba | Epoch Capabilities Index | 143.77 | Source |
| Qwen 3.6 Flash | Alibaba | Epoch Capabilities Index | 143.05 | Source |
| Gemini 2.5 Flash (Sep 2025) | Google DeepMind | Epoch Capabilities Index | 143 | Source |
| Gemma 4 31B IT | Google DeepMind | Epoch Capabilities Index | 142.86 | Source |
| Claude Opus 4 | Anthropic | Epoch Capabilities Index | 142.65 | Source |
| Qwen3.5-35B-A3B | Alibaba | Epoch Capabilities Index | 142.48 | Source |
| GPT-5.5 Instant | OpenAI | Epoch Capabilities Index | 142.45 | Source |
| Gemini 2.5 Pro (May 2025) | Google DeepMind | Epoch Capabilities Index | 142.44 | Source |
| Claude Haiku 4.5 | Anthropic | Epoch Capabilities Index | 142.4 | Source |
| Qwen3-Max | Alibaba | Epoch Capabilities Index | 142.34 | Source |
| o1 | OpenAI | Epoch Capabilities Index | 141.84 | Source |
| Gemma 4 26B A4B | Google DeepMind | Epoch Capabilities Index | 141.81 | Source |
| Claude Sonnet 4 | Anthropic | Epoch Capabilities Index | 141.67 | Source |
| Gemini 2.5 Flash (May 2025) | Google DeepMind | Epoch Capabilities Index | 141.51 | Source |
| Mistral Medium 3.5 | Mistral AI | Epoch Capabilities Index | 141.47 | Source |
| DeepSeek-R1 (May 2025) | DeepSeek | Epoch Capabilities Index | 141.27 | Source |
| Claude 3.7 Sonnet | Anthropic | Epoch Capabilities Index | 141.14 | Source |
| Gemini 2.5 Flash (Jun 2025) | Google DeepMind | Epoch Capabilities Index | 140.77 | Source |
| o3-mini | OpenAI | Epoch Capabilities Index | 140.33 | Source |
| Grok-3 mini | xAI | Epoch Capabilities Index | 140.33 | Source |
| Kimi K2 (Jul 2025) | Moonshot | Epoch Capabilities Index | 140.12 | Source |
| gpt-oss-120b | OpenAI | Epoch Capabilities Index | 140 | Source |
| Gemini 2.5 Flash (Apr 2025) | Google DeepMind | Epoch Capabilities Index | 139.94 | Source |
| DeepSeek-V3.1 | DeepSeek | Epoch Capabilities Index | 139.91 | Source |
| Qwen3-30B-A3B-Thinking (Jul 2025) | Alibaba | Epoch Capabilities Index | 139.61 | Source |
| Qwen3.5-9B | Alibaba | Epoch Capabilities Index | 139.43 | Source |
| Qwen3-235B-A22B | Alibaba | Epoch Capabilities Index | 139.35 | Source |
| GPT-5 nano | OpenAI | Epoch Capabilities Index | 139.33 | Source |
| DeepSeek-R1 | DeepSeek | Epoch Capabilities Index | 138.96 | Source |
| Qwen3-235B-A22B-Instruct (Jul 2025) | Alibaba | Epoch Capabilities Index | 138.92 | Source |
| Qwen3-32B | Alibaba | Epoch Capabilities Index | 138.49 | Source |
| Grok 3 | xAI | Epoch Capabilities Index | 138.32 | Source |
| Qwen3-14B | Alibaba | Epoch Capabilities Index | 138.22 | Source |
| gpt-oss-20b | OpenAI | Epoch Capabilities Index | 137.81 | Source |
| QwQ-32B | Alibaba | Epoch Capabilities Index | 137.58 | Source |
| DeepSeek-R1-Distill-Qwen-32B | DeepSeek | Epoch Capabilities Index | 137.42 | Source |
| Qwen3-30B-A3B-Instruct (Jul 2025) | Alibaba | Epoch Capabilities Index | 137.42 | Source |
| GPT-4.1 | OpenAI | Epoch Capabilities Index | 136.77 | Source |
| GPT-4.5 | OpenAI | Epoch Capabilities Index | 136.73 | Source |
| Qwen3-30B-A3B | Alibaba | Epoch Capabilities Index | 136.19 | Source |
| Qwen3-8B | Alibaba | Epoch Capabilities Index | 136.17 | Source |
| DeepSeek-V3 (Mar 2025) | DeepSeek | Epoch Capabilities Index | 135.93 | Source |
| o1-mini | OpenAI | Epoch Capabilities Index | 135.82 | Source |
| DeepSeek-R1-Distill-Qwen-14B | DeepSeek | Epoch Capabilities Index | 135.44 | Source |
| Gemini 2.0 Flash Thinking (Jan 2025) | Google DeepMind,Google | Epoch Capabilities Index | 135.36 | Source |
| Gemini 2.0 Pro | Google DeepMind | Epoch Capabilities Index | 135.05 | Source |
| GPT-4.1 mini | OpenAI | Epoch Capabilities Index | 135 | Source |
| o1-preview | OpenAI | Epoch Capabilities Index | 134.78 | Source |
| Gemini 2.0 Flash (Dec 2024) | Google DeepMind,Google | Epoch Capabilities Index | 134.69 | Source |
| Gemini 2.0 Flash (Feb 2025) | Google DeepMind,Google | Epoch Capabilities Index | 134.68 | Source |
| Mistral Medium 3 | Mistral AI | Epoch Capabilities Index | 134.06 | Source |
| Gemini 2.5 Flash-Lite (Jun 2025) | Google DeepMind | Epoch Capabilities Index | 133.95 | Source |
| Claude 3.5 Sonnet (October 2024) | Anthropic | Epoch Capabilities Index | 133.53 | Source |
| Magistral Small 1.0 | Mistral AI | Epoch Capabilities Index | 133.2 | Source |
| Qwen2.5-Max | Alibaba | Epoch Capabilities Index | 132.53 | Source |
| DeepSeek-V3 | DeepSeek | Epoch Capabilities Index | 132.33 | Source |
| Llama 4 Maverick | Meta AI | Epoch Capabilities Index | 132.2 | Source |
| Mistral Small 3.2 | Mistral AI | Epoch Capabilities Index | 131.75 | Source |
| Gemini 1.5 Pro (Sept 2024) | Google DeepMind | Epoch Capabilities Index | 131.73 | Source |
| Magistral Small 1.2 | Mistral AI | Epoch Capabilities Index | 131.43 | Source |
| Grok-2 (Dec 2024) | xAI | Epoch Capabilities Index | 130.49 | Source |
| Phi-4 | Microsoft Research | Epoch Capabilities Index | 130.43 | Source |
| Gemma 3 27B | Google DeepMind | Epoch Capabilities Index | 130.05 | Source |
| Claude 3.5 Sonnet | Anthropic | Epoch Capabilities Index | 130 | Source |
| Llama 4 Scout | Meta AI | Epoch Capabilities Index | 129.65 | Source |
| GPT-4.1 nano | OpenAI | Epoch Capabilities Index | 129.63 | Source |
| Gemini 1.5 Flash (Sep 2024) | Google DeepMind | Epoch Capabilities Index | 129.37 | Source |
| GPT-4o (May 2024) | OpenAI | Epoch Capabilities Index | 129.05 | Source |
| Qwen2.5-72B | Alibaba | Epoch Capabilities Index | 129 | Source |
| GPT-4o (Aug 2024) | OpenAI | Epoch Capabilities Index | 128.84 | Source |
| Llama 3.1-405B | Meta AI | Epoch Capabilities Index | 128.76 | Source |
| GPT-4o (Nov 2024) | OpenAI | Epoch Capabilities Index | 128.73 | Source |
| Qwen2.5-32B | Alibaba | Epoch Capabilities Index | 128.53 | Source |
| Mistral Large 2 (Nov 2024) | Mistral AI | Epoch Capabilities Index | 128.53 | Source |
| Mistral Large 2 (Jul 2024) | Mistral AI | Epoch Capabilities Index | 127.55 | Source |
| Mistral Small 3.1 | Mistral AI | Epoch Capabilities Index | 127.49 | Source |
| Llama 3.3 70B | Meta AI | Epoch Capabilities Index | 127.34 | Source |
| GPT-4 Turbo (Apr 2024) | OpenAI | Epoch Capabilities Index | 127.25 | Source |
| Claude 3.5 Haiku | Anthropic | Epoch Capabilities Index | 127.17 | Source |
| Mistral Small 3 | Mistral AI | Epoch Capabilities Index | 127.08 | Source |
| Gemini 1.5 Pro (May 2024) | Google DeepMind | Epoch Capabilities Index | 126.92 | Source |
| Claude 3 Opus | Anthropic | Epoch Capabilities Index | 126.91 | Source |
| GPT-4o mini | OpenAI | Epoch Capabilities Index | 126.58 | Source |
| GPT-4 Turbo (Nov 2023) | OpenAI | Epoch Capabilities Index | 126.46 | Source |
| Llama 3.1-70B | Meta AI | Epoch Capabilities Index | 125.94 | Source |
| GPT-4 (Mar 2023) | OpenAI | Epoch Capabilities Index | 125.88 | Source |
| Llama 3.2 90B | Meta AI | Epoch Capabilities Index | 125.51 | Source |
| Qwen2-72B | Alibaba | Epoch Capabilities Index | 125.28 | Source |
| DeepSeek-V2 (MoE-236B, May 2024) | DeepSeek | Epoch Capabilities Index | 124.78 | Source |
| Amazon Nova Pro | Amazon | Epoch Capabilities Index | 123.8 | Source |
| Gemma 3 12B | Google DeepMind | Epoch Capabilities Index | 123.49 | Source |
| GPT-4 (Jun 2023) | OpenAI | Epoch Capabilities Index | 123.16 | Source |
| Llama 3-70B | Meta AI | Epoch Capabilities Index | 122.94 | Source |
| Gemini 1.5 Flash (May 2024) | Google DeepMind | Epoch Capabilities Index | 122.61 | Source |
| Gemma 2 27B | Google DeepMind | Epoch Capabilities Index | 122.1 | Source |
| Mixtral 8x22B | Mistral AI | Epoch Capabilities Index | 122.05 | Source |
| Mistral Large | Mistral AI | Epoch Capabilities Index | 122.03 | Source |
| phi-3-small 7.4B | Microsoft | Epoch Capabilities Index | 121.85 | Source |
| phi-3-medium 14B | Microsoft | Epoch Capabilities Index | 121.25 | Source |
| Claude 3 Sonnet | Anthropic | Epoch Capabilities Index | 120.72 | Source |
| Claude Instant | Anthropic | Epoch Capabilities Index | 120.31 | Source |
| Claude 2 | Anthropic | Epoch Capabilities Index | 120.15 | Source |
| Gemma 2 9B | Google DeepMind | Epoch Capabilities Index | 119.85 | Source |
| Qwen2.5-Coder-32B | Alibaba | Epoch Capabilities Index | 119.52 | Source |
| Command R+ | Cohere,Cohere Labs (formerly Cohere for AI) | Epoch Capabilities Index | 119.37 | Source |
| Claude 2.1 | Anthropic | Epoch Capabilities Index | 119.3 | Source |
| Mistral NeMo | Mistral AI | Epoch Capabilities Index | 118.71 | Source |
| GPT-3.5 Turbo (Nov 2023) | OpenAI | Epoch Capabilities Index | 118.58 | Source |
| Qwen2.5-7B | Alibaba | Epoch Capabilities Index | 118.53 | Source |
| Mixtral 8x7B | Mistral AI | Epoch Capabilities Index | 118.5 | Source |
| Claude 3 Haiku | Anthropic | Epoch Capabilities Index | 118.39 | Source |
| Ministral 3B | Mistral AI | Epoch Capabilities Index | 118.12 | Source |
| Yi-34B | 01.AI | Epoch Capabilities Index | 117.43 | Source |
| phi-3-mini 3.8B | Microsoft | Epoch Capabilities Index | 117.4 | Source |
| Stable Beluga 2 | Stability AI | Epoch Capabilities Index | 117.16 | Source |
| Gemini 1.0 Pro | Google DeepMind | Epoch Capabilities Index | 117.07 | Source |
| Llama 3.1-8B | Meta AI | Epoch Capabilities Index | 116.61 | Source |
| Llama 3-8B | Meta AI | Epoch Capabilities Index | 116.49 | Source |
| Qwen2.5-Coder-14B | Not supplied | Epoch Capabilities Index | 116.44 | Source |
| Gemma 3 4B | Google DeepMind | Epoch Capabilities Index | 116.01 | Source |
| GPT-3.5 Turbo (Jan 2024) | OpenAI | Epoch Capabilities Index | 115.74 | Source |
| PaLM 2-L | Not supplied | Epoch Capabilities Index | 115.07 | Source |
| Llama 2-70B | Meta AI | Epoch Capabilities Index | 113.84 | Source |
| GPT-3.5 Turbo (Jun 2023) | OpenAI | Epoch Capabilities Index | 113.42 | Source |
| Qwen2.5-Coder (7B) | Alibaba | Epoch Capabilities Index | 113.2 | Source |
| Qwen-14B | Alibaba | Epoch Capabilities Index | 113.09 | Source |
| Mistral 7B v0.1 | Mistral AI | Epoch Capabilities Index | 112.27 | Source |
| Falcon-180B | Technology Innovation Institute | Epoch Capabilities Index | 112.19 | Source |
| internlm-20b | Not supplied | Epoch Capabilities Index | 112.15 | Source |
| Gemma 7B | Google DeepMind | Epoch Capabilities Index | 112.05 | Source |
| DeepSeek LLM 67B | DeepSeek | Epoch Capabilities Index | 110.47 | Source |
| LLaMA-65B | Meta AI | Epoch Capabilities Index | 110.23 | Source |
| Falcon 2 11B | Technology Innovation Institute | Epoch Capabilities Index | 109.63 | Source |
| Mistral 7B v0.3 | Mistral AI | Epoch Capabilities Index | 109.06 | Source |
| DeepSeek-Coder-V2-Lite-Base | Not supplied | Epoch Capabilities Index | 108.96 | Source |
| PaLM 2-M | Not supplied | Epoch Capabilities Index | 108.37 | Source |
| Phi-2 | Microsoft | Epoch Capabilities Index | 108.02 | Source |
| Qwen2.5-Coder-3B | Not supplied | Epoch Capabilities Index | 107.82 | Source |
| Nemotron-4 15B | Nvidia | Epoch Capabilities Index | 107.8 | Source |
| Yi-9B | Not supplied | Epoch Capabilities Index | 107.69 | Source |
| LLaMA-33B | Meta AI | Epoch Capabilities Index | 107.49 | Source |
| Qwen-7B | Alibaba | Epoch Capabilities Index | 106.91 | Source |
| PaLM 2-S | Not supplied | Epoch Capabilities Index | 106.26 | Source |
| Llama 2-13B | Meta AI | Epoch Capabilities Index | 106.25 | Source |
| Llama 2-34B | Meta AI | Epoch Capabilities Index | 105.29 | Source |
| StarCoder 2 15B | Hugging Face,ServiceNow,Nvidia,BigCode | Epoch Capabilities Index | 105.09 | Source |
| Yi 6B | 01.AI | Epoch Capabilities Index | 104.86 | Source |
| Falcon-40B | Technology Innovation Institute | Epoch Capabilities Index | 104.54 | Source |
| Baichuan2-13B | Baichuan | Epoch Capabilities Index | 103.29 | Source |
| Qwen2.5-Coder (1.5B) | Alibaba | Epoch Capabilities Index | 103.09 | Source |
| internlm-7b | Not supplied | Epoch Capabilities Index | 102.97 | Source |
| Llama 3.2 1B | Meta AI | Epoch Capabilities Index | 102.04 | Source |
| INTELLECT-1 | Prime Intellect,Hugging Face,Arcee AI | Epoch Capabilities Index | 100.97 | Source |
| MPT-30B | MosaicML | Epoch Capabilities Index | 100.76 | Source |
| LLaMA-13B | Meta AI | Epoch Capabilities Index | 100.69 | Source |
| Llama 2-7B | Meta AI | Epoch Capabilities Index | 99.18 | Source |
| chatglm2-6b | Not supplied | Epoch Capabilities Index | 99.05 | Source |
| LLaMA-7B | Meta AI | Epoch Capabilities Index | 96.82 | Source |
| Baichuan 2-7B | Baichuan | Epoch Capabilities Index | 96.45 | Source |
| DeepSeek Coder 33B | DeepSeek,Peking University | Epoch Capabilities Index | 96.45 | Source |
| Falcon-7B | Technology Innovation Institute | Epoch Capabilities Index | 95.23 | Source |
| CodeQwen1.5-7B | Not supplied | Epoch Capabilities Index | 94.99 | Source |
| vicuna-13b-v1.1 | Not supplied | Epoch Capabilities Index | 94.8 | Source |
| MPT-7B | MosaicML | Epoch Capabilities Index | 94.73 | Source |
| Gemma 2B | Google DeepMind | Epoch Capabilities Index | 94.34 | Source |
| StarCoder 2 7B | Hugging Face,ServiceNow,Nvidia,BigCode | Epoch Capabilities Index | 93.7 | Source |
| XGen-7B | Salesforce | Epoch Capabilities Index | 93.54 | Source |
| Qwen-1_8B | Not supplied | Epoch Capabilities Index | 93.07 | Source |
| open_llama_7b | Not supplied | Epoch Capabilities Index | 91.8 | Source |
| Phi-1.5 | Microsoft | Epoch Capabilities Index | 91.68 | Source |
| Baichuan1-7B | Baichuan | Epoch Capabilities Index | 90.61 | Source |
| RedPajama-INCITE-7B-Base | Not supplied | Epoch Capabilities Index | 90.58 | Source |
| Dolly 2.0-12b | Databricks | Epoch Capabilities Index | 89.83 | Source |
| DeepSeek Coder 6.7B | DeepSeek,Peking University | Epoch Capabilities Index | 89.77 | Source |
| StarCoder 2 3B | Hugging Face,ServiceNow,Nvidia,BigCode | Epoch Capabilities Index | 88.92 | Source |
| Qwen2.5-Coder-0.5B | Not supplied | Epoch Capabilities Index | 88.31 | Source |
| Cerebras-GPT-13B | Cerebras Systems | Epoch Capabilities Index | 83.42 | Source |
| DeepSeek Coder 1.3B | DeepSeek,Peking University | Epoch Capabilities Index | 63.88 | Source |
| stablelm-tuned-alpha-7b | Not supplied | Epoch Capabilities Index | 56.05 | Source |
No matching records. Try another model or organisation.
Measurement library
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
External benchmark
Epoch-administered evaluation. See source method and run settings; scores are not interchangeable across tasks.
A general capability result cannot tell us whether a tender deadline stayed current across four workers, or whether an approval refers to the document about to be sent.
Inspect our operational evaluations