加载中...
加载中...
Open LLM Leaderboard是追踪大模型评测结果的排行榜，通过追踪大语言模型和ChatBot在不同评测任务上的表现来对模型进行排名和评估。
数据来源: HuggingFace
| 模型名称 | 模型类型 | 参数大小(亿) | 平均分 | ARC分数 | HellaSwag分数 | MMLU分数 | TruthfulQA分数 | Winogrande分数 | GSM8K分数 | 模型架构 |
|---|---|---|---|---|---|---|---|---|---|---|
| Samantha-1.11-7b | Chat Models | 66.1 | 51.07 | 55.03 | 79.12 | 40.51 | 50.37 | 74.19 | 7.2 | Unknown |
| FuseLLM-7B | Pretrained Models | 70 | 51.07 | 53.24 | 78.72 |
数据仅供参考，以官方来源为准。模型名称旁的链接可跳转到 DataLearner 模型详情页。
| 47.93 |
| 38.17 |
| 74.03 |
| 14.33 |
| LlamaForCausalLM |
| deepseek-moe-16b-base | Pretrained Models | 163.8 | 51.07 | 53.24 | 79.77 | 46.31 | 36.08 | 73.72 | 17.29 | Unknown |
| Llama2-7b-sharegpt4 | Fine Tuned Models | 70 | 51.05 | 55.72 | 80.94 | 47.47 | 48.34 | 71.19 | 2.65 | LlamaForCausalLM |
| phi-2-OpenHermes-2.5 | Fine Tuned Models | 27.8 | 51.05 | 56.48 | 73.88 | 54.8 | 48.1 | 73.01 | 0.0 | PhiForCausalLM |
| chinese-llama-2-13b | Fine Tuned Models | 129.7 | 51.04 | 55.8 | 79.53 | 53.01 | 38.24 | 75.69 | 3.94 | Unknown |
| vicuna-7b-v1.3-sparsity-10 | Fine Tuned Models | 67.4 | 51.02 | 51.45 | 76.98 | 47.95 | 46.88 | 69.77 | 13.12 | LlamaForCausalLM |
| Llama-2-7b-chat-hf-30-sparsity | Chat Models | 67.4 | 51.02 | 52.47 | 76.58 | 45.57 | 44.82 | 69.61 | 17.06 | LlamaForCausalLM |
| yi6B_Vicuna | Fine Tuned Models | 60.6 | 51.02 | 46.16 | 69.3 | 58.43 | 48.11 | 65.67 | 18.42 | LlamaForCausalLM |
| Mistral-Pygmalion-7b | Fine Tuned Models | 70 | 51.02 | 54.44 | 78.48 | 49.23 | 41.82 | 75.3 | 6.82 | LlamaForCausalLM |
| Dorflan | Fine Tuned Models | 66.1 | 50.96 | 54.44 | 75.78 | 51.36 | 51.17 | 72.61 | 0.38 | Unknown |
| llama-2-7b-instruct-peft | Fine Tuned Models | 70 | 50.94 | 51.19 | 78.92 | 46.63 | 48.5 | 74.43 | 5.99 | Unknown |
| LLaMa-2-PeanutButter_v18_B-7B | Chat Models | 70 | 50.94 | 54.61 | 81.0 | 47.07 | 41.93 | 74.51 | 6.52 | Unknown |
| Llama-2-7b-chat-hf-afr-100step-v2 | Chat Models | 70 | 50.89 | 52.65 | 78.25 | 48.47 | 45.18 | 72.3 | 8.49 | LlamaForCausalLM |
| Barcenas-7b | Fine Tuned Models | 70 | 50.87 | 55.12 | 77.4 | 49.27 | 43.64 | 73.64 | 6.14 | LlamaForCausalLM |
| airoboros-l2-13b-2.1 | Fine Tuned Models | 130 | 50.84 | 55.12 | 80.24 | 50.89 | 44.62 | 71.9 | 2.27 | LlamaForCausalLM |
| Guanaco-Vicuna-7B-L2 | Fine Tuned Models | 66.1 | 50.83 | 53.24 | 78.89 | 46.77 | 42.75 | 75.37 | 7.96 | Unknown |
| GEITje-7B-chat-v2 | Chat Models | 72.4 | 50.79 | 50.34 | 74.13 | 49.0 | 43.55 | 71.51 | 16.22 | MistralForCausalLM |
| firefly-llama2-13b-pretrain | Fine Tuned Models | 129.7 | 50.77 | 53.92 | 79.1 | 51.25 | 36.24 | 75.53 | 8.57 | Unknown |
| LLaMa-2-PeanutButter_v10-7B | Chat Models | 70 | 50.75 | 55.29 | 81.69 | 46.97 | 43.78 | 70.88 | 5.91 | Unknown |
| vicuna-7b-v1.5-lora-timedial-unit-080082 | Fine Tuned Models | 66.1 | 50.74 | 52.82 | 76.07 | 50.47 | 43.54 | 73.72 | 7.81 | Unknown |
| starling-7B | Chat Models | 70 | 50.73 | 51.02 | 76.77 | 47.75 | 48.18 | 70.56 | 10.08 | LlamaForCausalLM |
| vicuna-7b-v1.5-lora-timedial-unit-080091 | Fine Tuned Models | 66.1 | 50.71 | 52.82 | 76.1 | 50.58 | 43.4 | 73.72 | 7.66 | Unknown |
| doctorLLM10k | Fine Tuned Models | 67.4 | 50.7 | 54.95 | 79.94 | 44.4 | 44.76 | 70.01 | 10.16 | LlamaForCausalLM |
| Llama-2-7b-chat-finetune-AUTOMATE | Fine Tuned Models | 70 | 50.68 | 53.07 | 75.59 | 48.8 | 44.73 | 73.24 | 8.64 | LlamaForCausalLM |
| vicuna-7b-v1.3-attention-sparsity-20 | Chat Models | 67.4 | 50.63 | 52.3 | 77.05 | 47.39 | 46.62 | 69.22 | 11.22 | LlamaForCausalLM |
| LaOT | Fine Tuned Models | 0 | 50.62 | 55.63 | 78.96 | 50.3 | 44.72 | 74.11 | 0.0 | Unknown |
| zaraxls-l2-7b | Fine Tuned Models | 70 | 50.61 | 54.44 | 78.94 | 50.39 | 46.51 | 73.16 | 0.23 | LlamaForCausalLM |
| llama-2-7b-guanaco-instruct-sharded | Chat Models | 67.4 | 50.58 | 53.75 | 78.69 | 46.65 | 43.93 | 72.61 | 7.81 | LlamaForCausalLM |
| llama-2-7b-miniguanaco | Fine Tuned Models | 67.4 | 50.55 | 49.06 | 75.59 | 46.14 | 43.73 | 72.61 | 16.15 | LlamaForCausalLM |
| vicuna-mmlu-val-mcq-7b-ep2 | Fine Tuned Models | 70 | 50.55 | 53.33 | 77.73 | 46.85 | 43.87 | 71.27 | 10.24 | LlamaForCausalLM |
| L2-7b-Guanaco-Uncensored | Fine Tuned Models | 66.1 | 50.55 | 50.6 | 76.99 | 48.93 | 43.42 | 75.37 | 7.96 | Unknown |
| GEITje-7B | Fine Tuned Models | 72.4 | 50.53 | 44.8 | 75.31 | 50.1 | 40.45 | 72.38 | 20.17 | MistralForCausalLM |
| monika-ddlc-7b-v1 | Fine Tuned Models | 70 | 50.49 | 54.95 | 76.78 | 45.61 | 43.94 | 72.85 | 8.79 | LlamaForCausalLM |
| WizardCoder-Python-34B-V1.0 | Chat Models | 340 | 50.46 | 52.13 | 74.78 | 49.15 | 48.85 | 68.35 | 9.48 | LlamaForCausalLM |
| openthaigpt-1.0.0-beta-13b-chat-hf | Fine Tuned Models | 130 | 50.45 | 53.58 | 79.09 | 51.13 | 44.16 | 73.88 | 0.83 | LlamaForCausalLM |
| MistralLite-11B | Fine Tuned Models | 107.3 | 50.43 | 57.68 | 79.54 | 50.09 | 38.27 | 76.64 | 0.38 | MistralForCausalLM |
| Koss-7B-chat | Fine Tuned Models | 70 | 50.37 | 53.67 | 78.79 | 46.72 | 43.97 | 71.74 | 7.35 | LlamaForCausalLM |
| vicuna-7b-1.1 | Fine Tuned Models | 70 | 50.37 | 53.67 | 77.46 | 45.63 | 48.94 | 70.96 | 5.53 | LlamaForCausalLM |
| spatial-vicuna-7b-v1.5-LoRA | Fine Tuned Models | 70 | 50.36 | 50.77 | 74.63 | 48.13 | 49.36 | 72.38 | 6.9 | Unknown |
| vicuna-7b-v1.5-lora-timedial | Fine Tuned Models | 66.1 | 50.35 | 52.9 | 76.29 | 50.47 | 41.6 | 73.56 | 7.28 | Unknown |
| Fireplace-13b | Fine Tuned Models | 130.2 | 50.34 | 47.7 | 69.61 | 43.56 | 48.24 | 67.17 | 25.78 | LlamaForCausalLM |
| vicuna-7b-v1.3-attention-sparsity-30 | Chat Models | 67.4 | 50.33 | 51.02 | 76.41 | 46.83 | 46.06 | 69.3 | 12.36 | LlamaForCausalLM |
| Yi-Ko-6B | Fine Tuned Models | 61.8 | 50.27 | 48.89 | 74.48 | 55.72 | 37.09 | 72.93 | 12.51 | LlamaForCausalLM |
| smol-3b | Fine Tuned Models | 30.2 | 50.27 | 46.33 | 68.23 | 46.33 | 50.73 | 65.35 | 24.64 | Unknown |
| Asclepius-Llama2-13B | Chat Models | 130 | 50.25 | 55.89 | 79.66 | 52.38 | 40.76 | 72.69 | 0.15 | LlamaForCausalLM |
| tulu-7B-fp16 | Fine Tuned Models | 70 | 50.24 | 50.17 | 77.04 | 47.63 | 41.61 | 73.8 | 11.22 | LlamaForCausalLM |
| MiniChat-1.5-3B | Fine Tuned Models | 30 | 50.23 | 46.5 | 68.28 | 46.67 | 50.71 | 65.04 | 24.18 | LlamaForCausalLM |
| LLongMA-2-13b-16k | Fine Tuned Models | 130 | 50.22 | 54.27 | 79.63 | 50.97 | 37.71 | 72.77 | 5.99 | Unknown |
| stack-llama-2 | Fine Tuned Models | 0 | 50.21 | 53.07 | 78.57 | 46.8 | 38.75 | 74.03 | 10.01 | LlamaForCausalLM |
| Llama-2-7b-chat-hf-afr-200step-v2 | Chat Models | 70 | 50.21 | 51.79 | 77.41 | 48.55 | 43.69 | 71.9 | 7.88 | LlamaForCausalLM |
| ALMA-13B | Fine Tuned Models | 130 | 50.16 | 56.83 | 80.29 | 49.92 | 37.57 | 76.32 | 0.0 | LlamaForCausalLM |
| LLongMA-2-13b-16k | Fine Tuned Models | 130 | 50.09 | 54.27 | 79.66 | 50.86 | 37.68 | 72.61 | 5.46 | Unknown |
| Llama-2-7B-32K-Instruct | Fine Tuned Models | 70 | 50.02 | 51.11 | 78.51 | 46.11 | 44.86 | 73.88 | 5.69 | LlamaForCausalLM |
| llama-7b-SFT-qlora-eli5-wiki_DPO_ds_RM_top_2_1024_r_64_alpha_16 | Chat Models | 70 | 49.98 | 54.1 | 78.74 | 45.44 | 43.4 | 73.64 | 4.55 | Unknown |
| Platypus2-7B | Chat Models | 67.4 | 49.97 | 55.2 | 78.84 | 49.83 | 40.64 | 73.48 | 1.82 | LlamaForCausalLM |
| belal-finetuned-llama2-1024-v2.2 | Fine Tuned Models | 67.4 | 49.96 | 52.65 | 77.81 | 44.65 | 40.02 | 74.11 | 10.54 | LlamaForCausalLM |
| mamba-gpt-7b | Fine Tuned Models | 70 | 49.96 | 51.19 | 75.4 | 47.47 | 42.06 | 71.67 | 11.98 | LlamaForCausalLM |
| Uncensored-Jordan-7B | Fine Tuned Models | 70 | 49.95 | 51.28 | 77.37 | 45.69 | 47.5 | 71.11 | 6.75 | LlamaForCausalLM |
| llama-2-coder-7b | Fine Tuned Models | 70 | 49.95 | 54.01 | 78.35 | 46.25 | 38.49 | 75.45 | 7.13 | LlamaForCausalLM |
| tora-code-34b-v1.0 | Fine Tuned Models | 340 | 49.92 | 50.26 | 75.48 | 46.65 | 39.62 | 67.72 | 19.79 | LlamaForCausalLM |
| openbuddy-deepseekcoder-33b-v16.1-32k | Fine Tuned Models | 334 | 49.91 | 45.05 | 60.79 | 43.24 | 44.49 | 62.19 | 43.67 | LlamaForCausalLM |
| LLaMa-2-PeanutButter_v18_A-7B | Chat Models | 70 | 49.88 | 53.16 | 78.11 | 45.54 | 40.37 | 74.9 | 7.2 | Unknown |
| kollama2-7b | Fine Tuned Models | 70 | 49.81 | 53.24 | 78.78 | 42.31 | 44.56 | 73.95 | 5.99 | LlamaForCausalLM |
| Solar-M-SakuraSolar-Mixed | Fine Tuned Models | 89.9 | 49.81 | 45.9 | 58.56 | 64.51 | 59.62 | 70.24 | 0.0 | LlamaForCausalLM |
| WizardMath-7B-V1.0 | Fine Tuned Models | 70 | 49.78 | 54.1 | 79.55 | 45.97 | 43.65 | 72.69 | 2.73 | LlamaForCausalLM |
| ELYZA-japanese-Llama-2-7b-instruct | Chat Models | 70 | 49.78 | 53.16 | 78.25 | 47.07 | 39.08 | 73.24 | 7.88 | LlamaForCausalLM |
| llama2-to-mistral-diff | Fine Tuned Models | 0 | 49.78 | 53.41 | 78.56 | 46.43 | 38.71 | 74.03 | 7.51 | Unknown |
| vicuna-7b-v1.5-lora-mixed-datasets-time-unit | Fine Tuned Models | 66.1 | 49.77 | 51.79 | 76.41 | 49.58 | 40.33 | 73.4 | 7.13 | Unknown |
| LongQLoRA-Llama2-7b-8k | Fine Tuned Models | 70 | 49.75 | 52.47 | 78.11 | 45.37 | 38.94 | 72.06 | 11.52 | LlamaForCausalLM |
| llama-2-7b-hf_open-platypus | Chat Models | 67.4 | 49.73 | 51.45 | 78.63 | 43.6 | 43.71 | 74.43 | 6.6 | LlamaForCausalLM |
| test-llama2-7b | Fine Tuned Models | 70 | 49.73 | 53.07 | 78.57 | 46.86 | 38.75 | 74.03 | 7.13 | Unknown |
| Starlight-7B | Fine Tuned Models | 70 | 49.73 | 53.07 | 78.57 | 46.8 | 38.75 | 74.03 | 7.13 | LlamaForCausalLM |
| ToolLLaMA-7b-LoRA | Fine Tuned Models | 70 | 49.72 | 52.99 | 78.62 | 46.87 | 38.67 | 74.35 | 6.82 | Unknown |
| vicuna-class-shishya-ac-hal-13b-ep3 | Fine Tuned Models | 130 | 49.7 | 48.46 | 80.78 | 56.17 | 39.32 | 73.48 | 0.0 | LlamaForCausalLM |
| vicuna-7b-v1.5-lora-mixed-datasets | Fine Tuned Models | 66.1 | 49.7 | 51.71 | 76.44 | 50.13 | 39.57 | 73.24 | 7.13 | Unknown |
| finetuned-llama-v2.0 | Fine Tuned Models | 0 | 49.67 | 53.16 | 77.75 | 43.69 | 39.08 | 74.43 | 9.93 | LlamaForCausalLM |
| llama2_7b_chat_uncensored | Fine Tuned Models | 70 | 49.67 | 53.58 | 78.66 | 44.49 | 41.34 | 74.11 | 5.84 | LlamaForCausalLM |
| chinese-alpaca-plus-13b-hf | Fine Tuned Models | 130 | 49.66 | 53.16 | 73.51 | 48.81 | 45.32 | 75.06 | 2.12 | LlamaForCausalLM |
| starchat-beta | Fine Tuned Models | 155.2 | 49.66 | 52.47 | 80.59 | 42.85 | 47.22 | 69.69 | 5.16 | GPTBigCodeForCausalLM |
| Llama-2-7B-32K-Instruct | Chat Models | 70 | 49.65 | 51.37 | 78.47 | 45.53 | 45.01 | 72.85 | 4.7 | LlamaForCausalLM |
| airoboros-l2-7b-2.1 | Fine Tuned Models | 70 | 49.64 | 54.44 | 78.68 | 44.45 | 43.95 | 74.11 | 2.2 | LlamaForCausalLM |
| Llama-2-7b-ft-instruct-es | Fine Tuned Models | 70 | 49.63 | 53.67 | 77.83 | 46.58 | 38.82 | 75.22 | 5.69 | LlamaForCausalLM |
| meditron-7b-chat | Fine Tuned Models | 67.4 | 49.59 | 50.77 | 75.37 | 40.49 | 48.56 | 73.16 | 9.17 | LlamaForCausalLM |
| airoboros-l2-7b-gpt4-1.4.1 | Fine Tuned Models | 70 | 49.54 | 55.12 | 79.6 | 45.17 | 40.29 | 74.27 | 2.81 | LlamaForCausalLM |
| llama2-ko-7B-model | Fine Tuned Models | 66.7 | 49.52 | 56.31 | 79.51 | 45.71 | 40.98 | 72.06 | 2.58 | Unknown |
| instruct-13b | Fine Tuned Models | 128.5 | 49.52 | 56.14 | 80.27 | 47.89 | 36.97 | 73.56 | 2.27 | Unknown |
| QuantumLM-7B | Unkown Model Types | 70 | 49.51 | 50.26 | 76.1 | 45.27 | 46.25 | 71.51 | 7.66 | LlamaForCausalLM |
| tamil-llama-13b-base-v0.1 | Fine Tuned Models | 130 | 49.5 | 52.82 | 79.95 | 52.05 | 36.56 | 75.61 | 0.0 | LlamaForCausalLM |
| AceGPT-7B | Fine Tuned Models | 70 | 49.47 | 53.58 | 77.54 | 43.0 | 38.75 | 72.77 | 11.14 | LlamaForCausalLM |
| Manticore-13B-Chat-Pyg-Guanaco-SuperHOT-8K-GPTQ | Unkown Model Types | 162.2 | 49.47 | 52.82 | 79.63 | 39.83 | 52.55 | 71.82 | 0.15 | LlamaForCausalLM |
| Guanaco-7B-Uncensored | Fine Tuned Models | 70 | 49.35 | 52.13 | 78.77 | 43.42 | 44.45 | 73.09 | 4.25 | LlamaForCausalLM |
| FinanceConnect-13B | Fine Tuned Models | 130.2 | 49.34 | 55.12 | 77.73 | 52.08 | 37.68 | 71.82 | 1.59 | LlamaForCausalLM |
| ALMA-13B-R | Chat Models | 130.2 | 49.32 | 55.55 | 79.45 | 49.52 | 36.09 | 75.3 | 0.0 | ? |
| odia_llama2_7B_base | Chat Models | 70 | 49.3 | 50.77 | 75.94 | 46.1 | 37.27 | 70.8 | 14.94 | LlamaForCausalLM |
| vigogne-7b-chat | Fine Tuned Models | 70 | 49.27 | 52.47 | 78.35 | 39.51 | 44.52 | 73.16 | 7.58 | LlamaForCausalLM |
| llama-v2-7b-32kC-Security | Chat Models | 66.1 | 49.19 | 49.83 | 77.33 | 44.41 | 47.96 | 71.74 | 3.87 | Unknown |
| ELYZA-japanese-Llama-2-7b-fast-instruct | Fine Tuned Models | 70 | 49.15 | 53.75 | 77.55 | 46.85 | 38.84 | 71.59 | 6.29 | LlamaForCausalLM |
| TowerBase-7B-v0.1 | Fine Tuned Models | 67.4 | 49.11 | 51.02 | 77.68 | 43.48 | 37.29 | 72.06 | 13.12 | LlamaForCausalLM |
| WhiteRabbitNeo-13B-v1 | Fine Tuned Models | 130 | 49.11 | 48.55 | 68.7 | 43.04 | 44.58 | 67.4 | 22.37 | LlamaForCausalLM |