Qwen2-72B-Instruct

Home/Models/Qwen/Qwen2-72B-Instruct

Qwen/Qwen2-72B-Instruct

kind: original
quant: bf16 (16-bit)
family: qwen
modality: text-generation
root model: Qwen/Qwen2-72B
provider: Qwen
license: other
language: en
context: 32768,
library: —
vRAM (GB): —
size on disk: —
downloads: 61,235
likes: 718
created: —
last modified: 2024-10-08T05:16:14+00:00 (2024)
huggingface: https://huggingface.co/Qwen/Qwen2-72B-Instruct
tags: transformerssafetensorsqwen2text-generationchatconversationalenarxiv:2309.00071base_model:Qwen/Qwen2-72Bbase_model:finetune:Qwen/Qwen2-72Blicense:othereval-resultstext-generation-inferenceendpoints_compatibleregion:us

benchmark	category	score	context	conf	evidence
GSM8K	math	93.0	GSM8K	medium	`\| GSM8K \| 93.0 \| 82.7 \| 91.1 \|`
MBPP	coding	82.3	MBPP	medium	`\| MBPP \| 82.3 \| 71.9 \| 80.2 \|`
MMLU	knowledge	82.0	MMLU	medium	`\| MMLU \| 82.0 \| 75.6 \| 82.3 \|`
HumanEval	coding	81.7	HumanEval	medium	`\| HumanEval \| 81.7 \| 71.3 \| 86.0 \|`
IFEval	instruction-following	77.3	IFEval (Prompt Strict-Acc.)	medium	`\| IFEval (Prompt Strict-Acc.) \| 77.3 \| 55.8 \| 77.6 \|`
EvalPlus	coding	75.2	EvalPlus	medium	`\| EvalPlus \| 75.2 \| 66.9 \| 79.0 \|`
C-Eval	knowledge	61.6	C-Eval	medium	`\| C-Eval \| 61.6 \| 76.1 \| 83.8 \|`
MMLU-Pro	knowledge	56.2	MMLU-Pro	medium	`\| MMLU-Pro \| 56.2 \| 51.7 \| 64.4 \|`
MATH	math	50.4	MATH	medium	`\| MATH \| 50.4 \| 42.5 \| 59.7 \|`
GPQA	reasoning	41.9	GPQA	medium	`\| GPQA \| 41.9 \| 39.4 \| 42.4 \|`
Chatbot Arena	chat	41.1	Arena-Hard	medium	`\| Arena-Hard \| 41.1 \| 36.1 \| 48.1 \|`
LiveCodeBench	coding	29.3	LiveCodeBench	medium	`\| LiveCodeBench \| 29.3 \| 17.9 \| 35.7 \|`
MT-Bench	chat	8.95	MT-Bench	medium	`\| MT-Bench \| 8.95 \| 8.61 \| 9.12 \|`

none scraped.