HF Model & Benchmark Matrix

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

DeepSeek-V2.5

deepseek-ai/DeepSeek-V2.5

kind
original
quant
bf16 (16-bit)
family
deepseek
modality
text-generation
root model
deepseek-ai/DeepSeek-V2.5
provider
deepseek-ai
license
other
language
—
context
—
library
transformers
vRAM (GB)
—
size on disk
—
downloads
4,430
likes
734
created
—
last modified
2024-12-11T12:08:06+00:00 (2024)
huggingface
https://huggingface.co/deepseek-ai/DeepSeek-V2.5
tags
transformerssafetensorsdeepseek_v2text-generationconversationalcustom_codearxiv:2405.04434license:othereval-resultstext-generation-inferenceendpoints_compatibleregion:us

Benchmark scores (7)

benchmarkcategoryscorecontextconfevidence
HumanEvalcoding84.5HumanEval pythonmedium| HumanEval python | 84.5 | 87.2 | 89 |
HumanEvalcoding73.8HumanEval Multimedium| HumanEval Multi | 73.8 | 74.8 | 73.8 |
Chatbot Arenachat68.3ArenaHardmedium| ArenaHard | 68.3 | 66.3 | 76.2 |
AlpacaEvalchat46.6AlpacaEval 2.0medium| AlpacaEval 2.0 | 46.6 | 44.5 | 50.5 |
LiveCodeBenchcoding36.6LiveCodeBench(01-09)medium| LiveCodeBench(01-09) | 36.6 | 39.7 | 41.8 |
MT-Benchchat8.85MT-Benchmedium| MT-Bench | 8.85 | 8.91 | 9.02 |
AlpacaEvalchat2.0mediumAlpacaEval 2.0

Discussions (0)

none scraped.

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