HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

Qwen-7B-Chat-AWQ

TheBloke/Qwen-7B-Chat-AWQ

kind
awq
quant
awq
family
qwen
modality
text-generation
root model
Qwen/Qwen-7B-Chat
provider
TheBloke
license
—
language
zh, en
context
—
library
—
vRAM (GB)
—
size on disk
—
downloads
62
likes
9
created
—
last modified
2023-11-21T11:10:39+00:00 (2023)
huggingface
https://huggingface.co/TheBloke/Qwen-7B-Chat-AWQ
tags
transformerssafetensorsqwentext-generationcustom_codezhenarxiv:2309.16609arxiv:2305.08322arxiv:2009.03300arxiv:2305.05280arxiv:2210.03629base_model:Qwen/Qwen-7B-Chatbase_model:quantized:Qwen/Qwen-7B-Chat4-bitawqregion:us

Benchmark scores (4)

benchmarkcategoryscorecontextconfevidence
C-Evalknowledge59.7low| BF16 | 55.8 | 59.7 | 50.3 | 37.2 |
MMLUknowledge55.8low| BF16 | 55.8 | 59.7 | 50.3 | 37.2 |
GSM8Kmath50.3low| BF16 | 55.8 | 59.7 | 50.3 | 37.2 |
HumanEvalcoding37.2low| BF16 | 55.8 | 59.7 | 50.3 | 37.2 |

Discussions (0)

none scraped.

Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.