midashenglm-spatial

mispeech/midashenglm-spatial

Audio Text To Text8.4B parameterstransformersLicense: apache-2.0
Tracked by OpenModelStats since November 5, 2026View on Hugging Face ↗
Downloads 30D
456
Total downloads
523
Gained 7D
—
Gained 30D
—
Likes
11
Rank · tracked models
#55,232
4330(up 4330 places)this week
Parameters
8.4B
Last source update
Oct 15, 2026
  • It moved from #70 to #69 among tracked Audio Text To Text models this week.

Downloads over time

Cumulative total downloads observed by OpenModelStats

Likes over time

Rank among tracked models

Lower is better

Overview

midashenglm-spatial is a audio text to text model published by mispeech. OpenModelStats has tracked the model since Nov 5, 2026, most recently observing it Nov 12, 2026.

Published
Sep 23, 2026
Last updated
Oct 15, 2026
Library
transformers
License
apache-2.0
Task
Audio Text To Text
Parameters
8,379,291,031
Spaces
—
Derivative models
—
Velocity 7D/day
—
transformerssafetensorsmidashenglm_spatialtext-generationspatial-audio-understandingmultimodalaudio-language-modelaudiodashengaudio-text-to-textcustom_codearxiv:2610.11156

Similar models

Models similar to midashenglm-spatial
ModelDL 30D
Qwen2-Audio-7B-InstructQwen132.9K
Qwen2-Audio-7BQwen8,686
Unified_Audio_Schematencent102
VibeVoice-ASR-HFmicrosoft14.8K
midashenglm-7b-0804-fp32mispeech5,848
midashenglm-7b-1021-bf16mispeech1,388
mispeech/midashenglm-spatial — Downloads, Growth & Stats | OpenModelStats