utLiquid AI · Downloadable
d1-Omni-2B
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Oct 24, 2026
- Parameters
- 2B
- 4-bit weights only
- ≈ 1 GB
- License
- other
WHAT’S NEW
New AI models.
Keep up without the noise.
The latest source listings from makers we track, covering language, images, video, audio, and 3D. Catalog checked Oct 28, 2026.
Sorted by the date a source repository was created, which may precede public release. This is a feed of new listings, not confirmed release announcements. Downloadable weights do not automatically mean an open-source license. Coverage and date meanings.
Audio listings
172 dated recordsinclusionAI · Downloadable
Ming-3.0-asr-tiny
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Oct 14, 2026
- Parameters
- 8.9B
- 4-bit weights only
- ≈ 4.4 GB
- License
- mit
Mistral AI · Downloadable
Voxtral-Mini-4B-Realtime-Arabic
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Oct 8, 2026
- Parameters
- 4B
- 4-bit weights only
- ≈ 2 GB
- License
- Not disclosed
Fish Audio · Downloadable
drama-3
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Oct 1, 2026
- Parameters
- 35.9B
- 4-bit weights only
- ≈ 17.9 GB
- License
- other
inclusionAI · Downloadable
Ming-3.0-tts-0.5B
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Sep 22, 2026
- Parameters
- 500M
- 4-bit weights only
- ≈ 0.3 GB
- License
- mit
inclusionAI · Downloadable
Ming-3.0-tts-tiny
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Sep 22, 2026
- Parameters
- 9B
- 4-bit weights only
- ≈ 4.5 GB
- License
- mit
Moondream · Downloadable
parakeet-ultra
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Sep 22, 2026
- Parameters
- 627.3M
- 4-bit weights only
- ≈ 0.3 GB
- License
- cc-by-4.0
Moondream · Downloadable
parakeet-redux
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Sep 18, 2026
- Parameters
- 148.9M
- 4-bit weights only
- ≈ 0.1 GB
- License
- cc-by-4.0
M-A-P · Downloadable
YuE2-3B
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Sep 9, 2026
- Parameters
- 3B
- 4-bit weights only
- ≈ 1.5 GB
- License
- cc-by-nc-4.0
inclusionAI · Downloadable
Ming-3.0-audio-flash
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Sep 3, 2026
- Parameters
- 128.2B
- 4-bit weights only
- ≈ 64.1 GB
- License
- mit
inclusionAI · Downloadable
Ming-3.0-audio-tiny
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Sep 3, 2026
- Parameters
- 8.9B
- 4-bit weights only
- ≈ 4.4 GB
- License
- mit
BAAI · Downloadable
GaussianMind
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Sep 3, 2026
- Parameters
- 5.6B
- 4-bit weights only
- ≈ 2.8 GB
- License
- apache-2.0
Microsoft · Downloadable
VibeVoice-ASR-Streaming-1.5B
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Sep 2, 2026
- Parameters
- 1.5B
- 4-bit weights only
- ≈ 0.8 GB
- License
- mit
Microsoft · Downloadable
VibeVoice-ASR-Streaming-7B
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Sep 2, 2026
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- mit
NVIDIA · Downloadable
NVIDIA-NemotronLabs-RE-USE-laughter
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Sep 1, 2026
- Parameters
- 9.6M
- 4-bit weights only
- ≈ 0 GB
- License
- openmdw-1.1
NVIDIA · Downloadable
nemotron-labs-unified-asr-0.6b
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Aug 21, 2026
- Parameters
- 600M
- 4-bit weights only
- ≈ 0.3 GB
- License
- other
Tencent · Downloadable
AuK-Flash
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Aug 21, 2026
- Parameters
- Not reported
- 4-bit weights only
- Not reported
- License
- mit
Tencent · Downloadable
AuK
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Aug 18, 2026
- Parameters
- Not reported
- 4-bit weights only
- Not reported
- License
- mit
MiniMax · Downloadable
MiniMax-Music3
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Aug 7, 2026
- Parameters
- 2.4B
- 4-bit weights only
- ≈ 1.2 GB
- License
- Not disclosed
Microsoft · Downloadable
VibeVoice-ASR-BitNet
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Jul 24, 2026
- Parameters
- 322.6M
- 4-bit weights only
- ≈ 0.2 GB
- License
- mit
NVIDIA · Downloadable
NVIDIA-NemotronLabs-RE-USE
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Jul 22, 2026
- Parameters
- 9.6M
- 4-bit weights only
- ≈ 0 GB
- License
- openmdw-1.1
OpenAI · Cloud
GPT-Realtime-2.1
Cloud API for vision, audio understanding, audio generation, chat & writing, reasoning, tool use. Capabilities are catalog-listed, not independently scored.
- Source listed
- Jul 6, 2026
- License
- Provider terms
OpenMOSS · Preview
MOSS-Transcribe-preview-2B
Works with speech or audio. Check the model card for transcription and generation capabilities.
- Source listed
- Jun 26, 2026
- Parameters
- 2B
- 4-bit weights only
- ≈ 1 GB
- License
- apache-2.0
Google · Preview
Gemini 3.5 Live Translate Preview
Cloud API for audio understanding, audio generation, chat & writing. Capabilities are catalog-listed, not independently scored.
- Source listed
- Jun 9, 2026
- License
- Provider terms
What makes this feed useful?
Capability tags distinguish a model that reads images from one that creates them. Each listing links to its source and a permanent profile, so you can inspect its license and compare it with a model you already know.
Does this cover every new release?
No. The current feed tracks public Hugging Face metadata from selected makers. It can miss untracked makers, cloud-only releases, and records beyond a source’s retrieval limit. Announced models without a source listing are not included. Check Sources & methodology for the current coverage.
Listings update with the catalog automatically. A missing or delayed source can leave older information visible; each model profile shows when it was checked.