HF Model & Benchmark Matrix

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

LFM2.5-Audio-1.5B-JP

LiquidAI/LFM2.5-Audio-1.5B-JP

kind
original
quant
bf16 (16-bit)
family
misc
modality
audio-to-audio
root model
LiquidAI/LFM2.5-Audio-1.5B
provider
LiquidAI
license
other
language
ja
context
—
library
liquid-audio
vRAM (GB)
—
size on disk
—
downloads
176
likes
68
created
—
last modified
2026-06-11T02:56:58+00:00 (2026)
huggingface
https://huggingface.co/LiquidAI/LFM2.5-Audio-1.5B-JP
tags
liquid-audiosafetensorsliquidlfm2audiolfm2-audiospeech-to-speechaudio-to-audiojaarxiv:2511.23404base_model:LiquidAI/LFM2.5-Audio-1.5Bbase_model:finetune:LiquidAI/LFM2.5-Audio-1.5Blicense:otherregion:us

Benchmark scores (2)

benchmarkcategoryscorecontextconfevidence
IFEvalinstruction-following14.3high| LFM2.5-Audio-1.5B-JP | 1.5B | 2.272 | **2.118** | 0.143 |
WER Common Voicespeech-asr4.42high| LFM2.5-Audio-1.5B-JP | **4.42** | 8.07 | 24.24 |

Discussions (0)

none scraped.

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