internlm2-math-plus-mixtral8x22b

Home/Models/internlm/internlm2-math-plus-mixtral8x22b

internlm/internlm2-math-plus-mixtral8x22b

kind: original
quant: bf16 (16-bit)
family: mistral
modality: text-generation
root model: internlm/internlm2-math-plus-mixtral8x22b
provider: internlm
license: other
language: en, zh
context: —
library: —
vRAM (GB): —
size on disk: —
downloads: 39
likes: 19
created: —
last modified: 2024-05-24T11:49:14+00:00 (2024)
huggingface: https://huggingface.co/internlm/internlm2-math-plus-mixtral8x22b
tags: transformerssafetensorsmixtraltext-generationmathenzharxiv:2402.06332license:othertext-generation-inferenceendpoints_compatibleregion:us

benchmark	category	score	context	conf	evidence
GSM8K	math	91.8		high	`\| InternLM2-Math-Plus-Mixtral8x22B \| 58.1 \| 68.5 \| 91.8 \|`
MATH	math	58.1		high	`\| InternLM2-Math-Plus-Mixtral8x22B \| 58.1 \| 68.5 \| 91.8 \|`
MATH	math	51.7	Deepseek-Math-7B-RL	medium	`\| Deepseek-Math-7B-RL \| 51.7 \| 58.8 \| 88.2 \|`
MATH	math	43.4	InternLM2-Math-Plus-7B	medium	`\| InternLM2-Math-Plus-7B \| 43.4 \|`
MATH	math	42.6	InternLM2-Math-Plus-20B	medium	`\| InternLM2-Math-Plus-20B \| 42.6 \|`
MATH	math	38.9	InternLM2-Math-Plus-1.8B	medium	`\| InternLM2-Math-Plus-1.8B \| 38.9 \|`
MATH	math	37.7	InternLM2-Math-20B	medium	`\| InternLM2-Math-20B \| 37.7 \| 54.3 \| 82.6 \|`
miniF2F	math	37.3		high	`\| InternLM2-Math-Plus-Mixtral8x22B \| 37.3 \|`
MATH	math	37.3	InternLM2-Math-Plus-Mixtral8x22B	medium	`\| InternLM2-Math-Plus-Mixtral8x22B \| 37.3 \|`
MATH	math	34.6	InternLM2-Math-7B	medium	`\| InternLM2-Math-7B \| 34.6 \| 50.9 \| 78.1 \|`
MATH	math	30.3	InternLM2-Math-7B-Base	medium	`\| InternLM2-Math-7B-Base \| 30.3 \|`
MATH	math	29.5	InternLM2-Math-20B-Base	medium	`\| InternLM2-Math-20B-Base \| 29.5 \|`

none scraped.