Epoch 能力榜(ECI)
按 Epoch AI 的综合能力指数排序,共 97 个模型。分数是原榜的。
这个榜怎么看?
- ECI(Epoch Capabilities Index):Epoch AI 自己把几十个基准测试合成的综合能力指数,数字越大越强。本站原样搬过来,不另外算分。
- ±:Epoch 给的区间。两个模型的区间重叠,差距可能只是误差。
- 只给最近一年上线的模型排名次;更早的在榜单页取消勾选就能看到。
名次后面的 ▲▼ 是和上一次更新相比的升降。
| # | 模型 | ECI | 上线 | 基准数 |
|---|---|---|---|---|
| 1▲1 | GPT-6 Astra OpenAI | ±4.3 | 2026-09-03 | 21 |
| 2▼1 | Claude Opus 5.5 Anthropic | ±4.0 | 2026-09-22 | 18 |
| 3 | GPT-6.1 Sol OpenAI | ±4.1 | 2026-09-29 | 18 |
| 4新 | GPT-6.1 Terra OpenAI | ±4.1 | 2026-10-26 | 10 |
| 5 | Gemini 4 Argon | ±3.6 | 2026-09-30 | 18 |
| 6▼2 | Claude Fable 5.1 Anthropic | ±3.5 | 2026-09-01 | 19 |
| 7▼1 | Claude Sonnet 5.5 Anthropic | ±3.7 | 2026-09-28 | 17 |
| 8▼1 | Claude Opus 5 Anthropic | ±3.1 | 2026-07-24 | 21 |
| 9▼1 | GPT-6 Sol OpenAI | ±3.1 | 2026-09-22 | 17 |
| 10▼1 | Claude Fable 5 Anthropic | ±3.4 | 2026-06-09 | 22 |
| 11▼1 | GPT-5.5 Pro OpenAI | ±3.3 | 2026-04-23 | 8 |
| 12▼1 | GPT-5.6 Sol OpenAI | ±3.1 | 2026-07-09 | 22 |
| 13▼1 | Muse Wonder Meta AI | ±2.8 | 2026-10-12 | 12 |
| 14▼1 | GPT-5.6 Terra OpenAI | ±2.8 | 2026-07-09 | 19 |
| 15 | GPT-5.4 Pro OpenAI | ±2.9 | 2026-03-05 | 9 |
| 16▼2 | GPT-5.5 OpenAI | ±2.8 | 2026-04-23 | 21 |
| 17▼1 | Claude Haiku 5.5 Anthropic | ±3.0 | 2026-10-07 | 18 |
| 18▼1 | Claude Opus 4.8 Anthropic | ±2.3 | 2026-05-28 | 21 |
| 19新 | DeepSeek V4.1 Flash 1017 DeepSeek | ±2.8 | 2026-10-17 | 11 |
| 20▼2 | Gemini 3.7 Flash | ±2.7 | 2026-08-13 | 16 |
| 21▼2 | Kimi K3 Moonshot | ±2.8 | 2026-07-16 | 20 |
| 22▼2 | Gemini 3.8 Flash | ±2.9 | 2026-09-02 | 19 |
| 23▼2 | Qwen 3.8 Max Alibaba | ±2.3 | 2026-08-02 | 12 |
| 24▼2 | GPT-5.4 OpenAI | ±2.1 | 2026-03-05 | 18 |
| 25▼2 | Muse Spark 1.3 Meta AI | ±2.2 | 2026-09-02 | 10 |
| 26▼2 | GPT-5.3 Codex OpenAI | ±3.7 | 2026-02-05 | 1 |
| 27▼2 | Claude Sonnet 5 Anthropic | ±2.5 | 2026-06-30 | 16 |
| 28 | Grok 4.6 xAI | ±2.5 | 2026-08-12 | 19 |
| 29▼2 | GPT-6 Luna OpenAI | ±2.7 | 2026-09-22 | 17 |
| 30▼4 | Claude Opus 4.7 Anthropic | ±2.1 | 2026-04-16 | 20 |
| 31▼2 | GPT-5.6 Luna OpenAI | ±2.7 | 2026-07-09 | 20 |
| 32▼2 | GPT-5.2 Pro OpenAI | ±3.2 | 2025-12-11 | 6 |
| 33▼2 | GLM-5.3 Z.ai | ±2.1 | 2026-08-14 | 13 |
| 34▼2 | DeepSeek V4.1 Flash DeepSeek | ±2.2 | 2026-09-09 | 17 |
| 35▼2 | Claude Opus 4.6 Anthropic | ±2.1 | 2026-02-05 | 20 |
| 36▼2 | DeepSeek V4 Pro 0813 DeepSeek | ±2.0 | 2026-08-13 | 16 |
| 37▼2 | Gemini 3.1 Pro | ±2.3 | 2026-02-19 | 20 |
| 38▼1 | Qwen3.8 Max (0902) Alibaba | ±2.0 | 2026-09-01 | 8 |
| 39▼1 | Muse Spark 1.2 Meta AI | ±2.3 | 2026-08-05 | 8 |
| 40▼4 | Muse Spark Open Meta AI | ±2.0 | 2026-10-15 | 8 |
| 41▼2 | Gemini 3.5 Flash | ±2.1 | 2026-05-19 | 17 |
| 42▼2 | DeepSeek V4 Flash 0731 DeepSeek | ±2.4 | 2026-07-31 | 15 |
| 43▼2 | Gemini 3.6 Flash | ±1.9 | 2026-07-21 | 16 |
| 44▼2 | Grok 4.7 xAI | ±2.2 | 2026-09-21 | 16 |
| 45▼2 | Grok 4.5 xAI | ±1.9 | 2026-07-08 | 18 |
| 46▼1 | Qwen3.7-Max Alibaba | ±2.2 | 2026-05-19 | 12 |
| 47▼3 | Muse Spark 1.1 Meta AI | ±2.5 | 2026-07-09 | 7 |
| 48▼2 | GPT-5.2 OpenAI | ±1.8 | 2025-12-11 | 18 |
| 49▼1 | Gemini 3 Pro | ±2.3 | 2025-11-18 | 9 |
| 50▼3 | GLM-5.3-Flash Z.ai | ±2.4 | 2026-08-20 | 12 |
| 51▼2 | Claude Sonnet 4.6 Anthropic | ±2.6 | 2026-02-17 | 13 |
| 52▼2 | Grok 4.20 xAI | ±2.5 | 2026-02-17 | 5 |
| 53▼2 | Muse Spark Meta AI | ±2.2 | 2026-04-08 | 3 |
| 54▼2 | GLM-5.2 Z.ai | ±2.1 | 2026-06-16 | 18 |
| 55▼2 | Gemini 3 Flash | ±1.7 | 2025-12-17 | 16 |
| 56▼2 | Qwen 3.8 27B Alibaba | ±1.8 | 2026-08-14 | 7 |
| 57▼2 | Kimi K2.6 Moonshot | ±1.8 | 2026-04-20 | 13 |
| — | GPT-5 Pro OpenAI | ±2.7 | 2025-10-07 | 7 |
| 58▲1 | Mistral Large 4 Mistral | ±3.1 | 2026-10-06 | 15 |
| 59▼3 | Inkling-Small Thinking Machines | ±2.1 | 2026-07-15 | 9 |
| 60新 | Beam 1.0 Reflection | ±1.8 | 2026-10-05 | 4 |
| 61▼4 | Claude Opus 4.5 Anthropic | ±2.6 | 2025-11-24 | 19 |
| 62▼4 | Kimi K2.7 Code Moonshot | ±2.0 | 2026-06-12 | 11 |
| — | GPT-5 OpenAI | 2025-08-07 | 18 | |
| 63▼3 | GLM-5.1 Z.ai | ±2.0 | 2026-04-07 | 9 |
| 64▼3 | GPT-5.1 OpenAI | ±1.6 | 2025-11-13 | 12 |
| 65▼3 | Qwen 3.6 Max (Preview) Alibaba | ±2.7 | 2026-04-20 | 7 |
| 66▼3 | DeepSeek-V4-Pro DeepSeek | ±1.6 | 2026-04-24 | 13 |
| 67▼3 | Grok 4.3 Beta xAI | ±1.6 | 2026-04-17 | 10 |
| 68▼3 | GPT-5.4 Mini OpenAI | ±1.8 | 2026-03-17 | 14 |
| 69▼3 | Inkling Thinking Machines | ±2.5 | 2026-07-15 | 16 |
| 70▼3 | Kimi K2.5 Moonshot | ±1.6 | 2026-01-27 | 5 |
| — | o3-pro OpenAI | ±2.2 | 2025-06-10 | 5 |
| 71▼3 | Qwen 3.6 Plus Alibaba | ±2.2 | 2026-03-31 | 8 |
| 72▼3 | Qwen3.7-Plus Alibaba | ±1.6 | 2026-06-02 | 7 |
| 73▼3 | MiniMax-M3 MiniMax | ±3.8 | 2026-06-01 | 11 |
| — | Claude Sonnet 4.5 Anthropic | ±1.7 | 2025-09-29 | 16 |
| 74▼3 | Qwen3.5 397B-A17B Alibaba | ±1.8 | 2026-02-13 | 7 |
| — | o3 OpenAI | ±1.8 | 2025-04-16 | 13 |
| 75▼3 | MiniMax-M2.5 MiniMax | ±3.4 | 2026-02-12 | 3 |
| — | Grok 4 xAI | ±1.8 | 2025-07-09 | 0 |
| 76▼3 | Qwen3.6 27B Alibaba | ±1.6 | 2026-04-22 | 6 |
| 77▼3 | Qwen 3.5 Plus (hosted 397B-A17B) Alibaba | ±2.1 | 2026-02-16 | 7 |
| 78▼3 | DeepSeek-V4-Flash DeepSeek | ±2.0 | 2026-04-24 | 4 |
| 79▼3 | DeepSeek-V3.2 DeepSeek | ±1.5 | 2025-12-01 | 5 |
| 80▼3 | GPT-5.4 Nano OpenAI | ±2.1 | 2026-03-17 | 13 |
| 81▼3 | Nemotron 3 Ultra Nvidia | ±2.3 | 2026-06-04 | 9 |
| 82▼2 | GLM-5 Z.ai | ±2.0 | 2026-02-11 | 6 |
| 83▼4 | Kimi K2 Thinking Moonshot | ±2.1 | 2025-11-06 | 1 |
| 84▼3 | MiniMax-M2.7 MiniMax | ±4.8 | 2026-03-18 | 2 |
| — | o4-mini OpenAI | ±2.1 | 2025-04-16 | 14 |
| — | GPT-5 mini OpenAI | ±1.8 | 2025-08-07 | 13 |
| — | Gemini 2.5 Pro (Jun 2025) | ±1.6 | 2025-06-05 | 0 |
| — | DeepSeek-V3.2-Exp DeepSeek | ±2.2 | 2025-09-29 | 3 |
| 85▼3 | Gemini 3.5 Flash-Lite | ±2.0 | 2026-07-21 | 13 |
| 86▼3 | Gemini 3.1 Flash-Lite | ±1.7 | 2026-03-03 | 6 |
| — | Grok 4 Fast xAI | ±2.1 | 2025-09-19 | 4 |
| — | Claude Opus 4.1 Anthropic | ±2.3 | 2025-08-05 | 10 |
| 87▼3 | Qwen3.7 Flash Alibaba | ±2.1 | 2026-07-27 | 4 |
| — | Gemini 2.5 Pro (Mar 2025) | ±2.6 | 2025-03-31 | 0 |
| — | Qwen3-235B-A22B-Thinking (Jul 2025) Alibaba | ±1.7 | 2025-07-25 | 0 |
| 88▼3 | Qwen 3.6 35B-A3B Alibaba | ±2.4 | 2026-04-14 | 8 |
| 89▼3 | GLM-4.7 Z.ai | ±2.2 | 2025-12-22 | 6 |
| 90▼3 | Qwen 3.5 Flash (hosted 35B-A3B) Alibaba | ±1.8 | 2026-02-25 | 8 |
| 91▼3 | Gemma 4 31B IT | ±2.7 | 2026-04-02 | 7 |
| — | Gemini 2.5 Flash (Sep 2025) | ±3.6 | 2025-09-25 | 0 |
| 92▼3 | Qwen 3.6 Flash Alibaba | ±2.2 | 2026-04-27 | 9 |
| — | Claude Opus 4 Anthropic | ±1.9 | 2025-05-22 | 9 |
| 93▼3 | Qwen3.5-35B-A3B Alibaba | ±2.3 | 2026-02-24 | 4 |
| 94▼3 | GPT-5.5 Instant OpenAI | ±2.7 | 2026-05-05 | 4 |
| — | Qwen3-Max Alibaba | ±2.3 | 2025-09-24 | 7 |
| — | Claude Haiku 4.5 Anthropic | ±2.3 | 2025-10-15 | 10 |
| — | Gemini 2.5 Pro (May 2025) | ±3.2 | 2025-05-06 | 0 |
| 95▼3 | Gemma 4 26B A4B | ±2.3 | 2026-04-02 | 3 |
| — | o1 OpenAI | ±1.5 | 2024-12-17 | 10 |
| — | Claude Sonnet 4 Anthropic | ±2.0 | 2025-05-22 | 9 |
| — | Gemini 2.5 Flash (May 2025) | ±1.8 | 2025-05-20 | 0 |
| 96▼3 | Mistral Medium 3.5 Mistral | ±2.5 | 2026-04-28 | 0 |
| — | DeepSeek-R1 (May 2025) DeepSeek | ±1.8 | 2025-05-28 | 0 |
| — | Claude 3.7 Sonnet Anthropic | ±1.7 | 2025-02-24 | 6 |
| — | Gemini 2.5 Flash (Jun 2025) | ±1.9 | 2025-06-17 | 0 |
| — | Grok-3 mini xAI | ±2.1 | 2025-06-24 | 5 |
| — | o3-mini OpenAI | ±2.2 | 2025-01-31 | 14 |
| — | Kimi K2 (Jul 2025) Moonshot | ±2.6 | 2025-07-12 | 0 |
| — | Gemini 2.5 Flash (Apr 2025) | ±3.3 | 2025-04-17 | 0 |
| — | DeepSeek-V3.1 DeepSeek | ±3.7 | 2025-08-21 | 4 |
| — | gpt-oss-120b OpenAI | ±3.6 | 2025-08-05 | 9 |
| — | Qwen3-30B-A3B-Thinking (Jul 2025) Alibaba | ±2.8 | 2025-07-30 | 0 |
| 97▼3 | Qwen3.5-9B Alibaba | ±2.4 | 2026-02-24 | 4 |
| — | GPT-5 nano OpenAI | ±3.2 | 2025-08-07 | 13 |
| — | Qwen3-235B-A22B Alibaba | ±2.7 | 2025-04-28 | 5 |
| — | DeepSeek-R1 DeepSeek | ±1.9 | 2025-01-20 | 7 |
| — | Qwen3-235B-A22B-Instruct (Jul 2025) Alibaba | ±2.5 | 2025-07-25 | 0 |
| — | Qwen3-32B Alibaba | ±2.5 | 2025-04-29 | 4 |
| — | Grok 3 xAI | ±2.1 | 2025-04-09 | 7 |
| — | Qwen3-14B Alibaba | ±3.2 | 2025-04-29 | 4 |
| — | gpt-oss-20b OpenAI | ±3.2 | 2025-08-05 | 5 |
| — | QwQ-32B Alibaba | ±4.6 | 2025-03-05 | 2 |
| — | Qwen3-30B-A3B-Instruct (Jul 2025) Alibaba | ±3.7 | 2025-07-29 | 0 |
| — | GPT-4.5 OpenAI | ±2.1 | 2025-02-27 | 0 |
| — | GPT-4.1 OpenAI | ±2.2 | 2025-04-14 | 11 |
| — | Qwen3-30B-A3B Alibaba | ±3.9 | 2025-04-29 | 5 |
| — | Qwen3-8B Alibaba | ±3.9 | 2025-04-28 | 4 |
| — | DeepSeek-V3 (Mar 2025) DeepSeek | ±2.5 | 2025-03-24 | 0 |
| — | o1-mini OpenAI | ±2.8 | 2024-09-12 | 6 |
| — | Gemini 2.0 Flash Thinking (Jan 2025) | ±3.8 | 2025-01-21 | 0 |
| — | Gemini 2.0 Pro | ±2.5 | 2025-02-05 | 0 |
| — | GPT-4.1 mini OpenAI | ±2.7 | 2025-04-14 | 11 |
| — | Gemini 2.0 Flash (Feb 2025) | ±2.5 | 2025-02-05 | 0 |
| — | Gemini 2.0 Flash (Dec 2024) | ±6.1 | 2024-12-11 | 0 |
| — | o1-preview OpenAI | ±3.3 | 2024-09-12 | 5 |
| — | Mistral Medium 3 Mistral | ±2.4 | 2025-05-07 | 0 |
| — | Gemini 2.5 Flash-Lite (Jun 2025) | ±3.0 | 2025-06-17 | 0 |
| — | Claude 3.5 Sonnet (October 2024) Anthropic | ±4.1 | 2024-10-22 | 0 |
| — | Magistral Small 1.0 Mistral | ±2.6 | 2025-06-10 | 0 |
| — | Qwen2.5-Max Alibaba | ±3.3 | 2025-01-25 | 0 |
| — | DeepSeek-V3 DeepSeek | ±3.3 | 2024-12-26 | 3 |
| — | Llama 4 Maverick Meta AI | ±3.0 | 2025-04-06 | 0 |
| — | Gemini 1.5 Pro (Sept 2024) | ±2.6 | 2024-09-24 | 0 |
| — | Phi-4 Microsoft Research | ±3.5 | 2024-12-12 | 3 |
| — | Grok-2 (Dec 2024) xAI | ±2.9 | 2024-12-12 | 0 |
| — | Claude 3.5 Sonnet Anthropic | 2024-06-20 | 7 | |
| — | Gemma 3 27B | ±3.8 | 2025-03-12 | 0 |
| — | Llama 4 Scout Meta AI | ±3.2 | 2025-04-05 | 0 |
| — | GPT-4.1 nano OpenAI | ±4.1 | 2025-04-14 | 9 |
| — | Gemini 1.5 Flash (Sep 2024) | ±4.0 | 2024-09-24 | 0 |
| — | GPT-4o (May 2024) OpenAI | ±3.2 | 2024-05-13 | 0 |
| — | Qwen2.5-72B Alibaba | ±3.4 | 2024-09-19 | 0 |
| — | GPT-4o (Aug 2024) OpenAI | ±3.1 | 2024-08-06 | 0 |
| — | Llama 3.1-405B Meta AI | ±3.1 | 2024-07-23 | 0 |
| — | GPT-4o (Nov 2024) OpenAI | ±3.5 | 2024-11-20 | 0 |
| — | Mistral Large 2 (Nov 2024) Mistral | ±3.2 | 2024-11-18 | 0 |
| — | Mistral Large 2 (Jul 2024) Mistral | ±3.6 | 2024-07-24 | 0 |
| — | Mistral Small 3.1 Mistral | ±3.4 | 2025-03-17 | 0 |
| — | Llama 3.3 70B Meta AI | ±3.5 | 2024-12-06 | 0 |
| — | GPT-4 Turbo (Apr 2024) OpenAI | ±3.2 | 2024-04-09 | 0 |
| — | Claude 3.5 Haiku Anthropic | ±4.3 | 2024-10-22 | 5 |
| — | Gemini 1.5 Pro (May 2024) | ±3.4 | 2024-05-14 | 0 |
| — | Claude 3 Opus Anthropic | ±3.5 | 2024-02-29 | 8 |
| — | GPT-4o mini OpenAI | ±3.9 | 2024-07-18 | 12 |
| — | GPT-4 Turbo (Nov 2023) OpenAI | ±3.8 | 2024-01-25 | 0 |
| — | Llama 3.1-70B Meta AI | ±3.7 | 2024-07-23 | 0 |
| — | GPT-4 (Mar 2023) OpenAI | ±5.1 | 2023-03-14 | 0 |
| — | Llama 3.2 90B Meta AI | ±3.7 | 2024-09-24 | 0 |
| — | Qwen2-72B Alibaba | ±3.7 | 2024-06-07 | 0 |
| — | DeepSeek-V2 (MoE-236B, May 2024) DeepSeek | ±4.4 | 2024-05-07 | 0 |
| — | Gemma 3 12B | ±6.3 | 2025-03-12 | 0 |
| — | GPT-4 (Jun 2023) OpenAI | ±4.7 | 2023-06-13 | 0 |
| — | Llama 3-70B Meta AI | ±4.1 | 2024-04-18 | 0 |
| — | Gemini 1.5 Flash (May 2024) | ±3.8 | 2024-05-23 | 0 |
| — | Mistral Large Mistral | ±4.7 | 2024-02-26 | 0 |
| — | Gemma 2 27B | ±4.2 | 2024-06-24 | 2 |
| — | Mixtral 8x22B Mistral | ±4.4 | 2024-04-17 | 0 |
| — | phi-3-small 7.4B Microsoft | ±4.6 | 2024-04-23 | 0 |
| — | phi-3-medium 14B Microsoft | ±4.4 | 2024-04-23 | 0 |
| — | Claude 3 Sonnet Anthropic | ±5.2 | 2024-02-29 | 4 |
| — | Claude Instant Anthropic | ±5.3 | 2023-08-09 | 0 |
| — | Claude 2 Anthropic | ±4.6 | 2023-07-11 | 0 |
| — | Gemma 2 9B | ±5.2 | 2024-06-24 | 0 |
| — | Qwen2.5-Coder-32B Alibaba | ±7.6 | 2024-09-18 | 0 |
| — | Claude 2.1 Anthropic | ±4.7 | 2023-11-21 | 4 |
| — | Command R+ Cohere | ±6.8 | 2024-08-30 | 0 |
| — | Claude 3 Haiku Anthropic | ±4.7 | 2024-03-07 | 5 |
| — | Mistral NeMo Mistral | ±4.9 | 2024-07-18 | 0 |
| — | Qwen2.5-7B Alibaba | ±5.2 | 2024-09-19 | 0 |
| — | GPT-3.5 Turbo (Nov 2023) OpenAI | ±5.9 | 2023-11-06 | 0 |
| — | Mixtral 8x7B Mistral | ±5.2 | 2023-12-11 | 0 |
| — | Ministral 3B Mistral | ±7.3 | 2024-10-16 | 0 |
| — | Yi-34B 01.AI | ±5.2 | 2023-11-02 | 0 |
| — | phi-3-mini 3.8B Microsoft | ±5.7 | 2024-04-23 | 0 |
| — | Gemini 1.0 Pro | ±4.6 | 2023-12-13 | 0 |
| — | Stable Beluga 2 Stability AI | ±5.4 | 2023-07-20 | 0 |
| — | Llama 3.1-8B Meta AI | ±7.9 | 2024-07-23 | 0 |
| — | Llama 3-8B Meta AI | ±5.1 | 2024-04-18 | 0 |
| — | Qwen2.5-Coder-14B 未知 | ±8.0 | 2024-09-18 | 0 |
| — | Gemma 3 4B | ±12.4 | 2025-03-12 | 0 |
| — | GPT-3.5 Turbo (Jan 2024) OpenAI | ±5.8 | 2024-01-25 | 0 |
| — | PaLM 2-L 未知 | ±10.3 | 2023-05-17 | 0 |
| — | Llama 2-70B Meta AI | ±5.5 | 2023-07-18 | 0 |
| — | GPT-3.5 Turbo (Jun 2023) OpenAI | ±6.0 | 2023-06-13 | 0 |
| — | Qwen2.5-Coder (7B) Alibaba | ±7.9 | 2024-09-18 | 0 |
| — | Qwen-14B Alibaba | ±6.4 | 2023-09-24 | 0 |
| — | Mistral 7B v0.1 Mistral | ±5.8 | 2023-09-27 | 0 |
| — | Falcon-180B Technology Innovation Institute | ±7.9 | 2023-09-06 | 0 |
| — | internlm-20b 未知 | ±6.5 | 2023-09-18 | 0 |
| — | Gemma 7B | ±6.9 | 2024-02-21 | 0 |
| — | LLaMA-65B Meta AI | ±6.2 | 2023-02-24 | 0 |
| — | Falcon 2 11B Technology Innovation Institute | ±6.5 | 2024-05-09 | 0 |
| — | Mistral 7B v0.3 Mistral | ±7.4 | 2024-05-27 | 0 |
| — | DeepSeek-Coder-V2-Lite-Base 未知 | ±7.5 | 2024-06-13 | 0 |
| — | PaLM 2-M 未知 | ±10.1 | 2023-05-17 | 0 |
| — | Phi-2 Microsoft | ±13.4 | 2023-12-12 | 0 |
| — | Qwen2.5-Coder-3B 未知 | ±10.9 | 2024-09-18 | 0 |
| — | Nemotron-4 15B Nvidia | ±7.2 | 2024-02-26 | 0 |
| — | Yi-9B 未知 | ±8.8 | 2024-03-01 | 0 |
| — | LLaMA-33B Meta AI | ±6.8 | 2023-02-24 | 0 |
| — | Qwen-7B Alibaba | ±9.0 | 2023-09-28 | 0 |
| — | PaLM 2-S 未知 | ±7.9 | 2023-05-17 | 0 |
| — | Llama 2-13B Meta AI | ±7.4 | 2023-07-18 | 0 |
| — | Llama 2-34B Meta AI | ±7.4 | 2023-07-18 | 0 |
| — | StarCoder 2 15B Hugging Face | ±8.8 | 2024-02-20 | 0 |
| — | Yi 6B 01.AI | ±7.9 | 2023-11-22 | 0 |
| — | Falcon-40B Technology Innovation Institute | ±7.5 | 2023-05-25 | 0 |
| — | Baichuan2-13B Baichuan | ±9.5 | 2023-09-06 | 0 |
| — | Qwen2.5-Coder (1.5B) Alibaba | ±11.9 | 2024-09-18 | 0 |
| — | internlm-7b 未知 | ±8.5 | 2023-07-05 | 0 |
| — | INTELLECT-1 Prime Intellect | ±7.9 | 2024-11-29 | 0 |
| — | MPT-30B MosaicML | ±7.6 | 2023-06-22 | 0 |
| — | LLaMA-13B Meta AI | ±8.2 | 2023-02-24 | 0 |
| — | Llama 2-7B Meta AI | ±8.5 | 2023-07-18 | 0 |
| — | chatglm2-6b 未知 | ±14.6 | 2023-06-24 | 0 |
| — | LLaMA-7B Meta AI | ±9.4 | 2023-02-24 | 0 |
| — | Baichuan 2-7B Baichuan | ±10.1 | 2023-09-20 | 0 |
| — | DeepSeek Coder 33B DeepSeek | ±9.6 | 2023-11-02 | 0 |
| — | Falcon-7B Technology Innovation Institute | ±9.9 | 2023-04-24 | 0 |
| — | CodeQwen1.5-7B 未知 | ±11.3 | 2024-04-15 | 0 |
| — | vicuna-13b-v1.1 未知 | ±11.1 | 2023-04-12 | 0 |
| — | MPT-7B MosaicML | ±9.4 | 2023-05-05 | 0 |
| — | Gemma 2B | ±9.5 | 2024-02-21 | 0 |
| — | StarCoder 2 7B Hugging Face | ±12.2 | 2024-02-20 | 0 |
| — | XGen-7B Salesforce | ±9.5 | 2023-06-27 | 0 |
| — | Qwen-1_8B 未知 | ±14.2 | 2023-11-30 | 0 |
| — | open_llama_7b 未知 | ±10.4 | 2023-06-07 | 0 |
| — | Phi-1.5 Microsoft | ±15.8 | 2023-09-11 | 0 |
| — | Baichuan1-7B Baichuan | ±10.8 | 2023-06-01 | 0 |
| — | RedPajama-INCITE-7B-Base 未知 | ±10.9 | 2023-05-04 | 0 |
| — | Dolly 2.0-12b Databricks | ±11.2 | 2023-04-11 | 0 |
| — | DeepSeek Coder 6.7B DeepSeek | ±12.1 | 2023-11-02 | 0 |
| — | StarCoder 2 3B Hugging Face | ±12.0 | 2024-02-22 | 0 |
| — | Qwen2.5-Coder-0.5B 未知 | ±20.3 | 2024-09-18 | 0 |
| — | Cerebras-GPT-13B Cerebras Systems | ±11.6 | 2023-03-20 | 0 |
| — | DeepSeek Coder 1.3B DeepSeek | ±16.1 | 2023-11-02 | 0 |
| — | stablelm-tuned-alpha-7b 未知 | ±21.4 | 2023-04-19 | 0 |