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 Nov 1, 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.

Latest listings

3,185 dated records

Google · Cloud

Gemini 2.5 Flash-Lite

VisionVideo understandingAudio understandingReasoningTool use

Cloud API for vision, video understanding, audio understanding, chat & writing, reasoning, tool use. Capabilities are catalog-listed, not independently scored.

Source listed: Jun 17, 2025
License: Provider terms

Compare this model →

Google · Cloud

Gemini 2.5 Pro

VisionVideo understandingAudio understandingReasoningTool use

Cloud API for vision, video understanding, audio understanding, chat & writing, reasoning, tool use. Capabilities are catalog-listed, not independently scored.

Source listed: Jun 17, 2025
License: Provider terms

Compare this model →

Moonshot AI · Downloadable

Kimi-Dev-72B

Text generation

Generates text for conversations and writing.

Source listed: Jun 16, 2025
Parameters: 72B
4-bit weights only: ≈ 36 GB
License: mit

Compare this model →

OpenBMB · Downloadable

Ultra-FineWeb-classifier

Capabilities unconfirmed

A specialized research model. See the model card for its intended task and supported setup.

Source listed: Jun 16, 2025
Parameters: Not reported
4-bit weights only: Not reported
License: apache-2.0

Compare this model →

k2-fsa · Downloadable

ZipVoice

Audio generation

Works with speech or audio. Check the model card for transcription and generation capabilities.

Source listed: Jun 15, 2025
Parameters: Not reported
4-bit weights only: Not reported
License: Not disclosed

Compare this model →

Google · Downloadable

videoprism-base-f16r288

Video understanding

Generates video. Resolution, clip length, and speed depend on the version and setup.

Source listed: Jun 14, 2025
Parameters: Not reported
4-bit weights only: Not reported
License: apache-2.0

Compare this model →

Meta · Downloadable

vjepa2-vitg-fpc64-384-ssv2

Video understanding

Generates video. Resolution, clip length, and speed depend on the version and setup.

Source listed: Jun 13, 2025
Parameters: 1.1B
4-bit weights only: ≈ 0.6 GB
License: mit

Compare this model →

Meta · Downloadable

vjepa2-vitl-fpc16-256-ssv2

Video understanding

Generates video. Resolution, clip length, and speed depend on the version and setup.

Source listed: Jun 13, 2025
Parameters: 375.5M
4-bit weights only: ≈ 0.2 GB
License: mit

Compare this model →

Meta · Downloadable

vjepa2-vitl-fpc32-256-diving48

Video understanding

Generates video. Resolution, clip length, and speed depend on the version and setup.

Source listed: Jun 13, 2025
Parameters: 375.4M
4-bit weights only: ≈ 0.2 GB
License: mit

Compare this model →

Meta · Downloadable

vjepa2-vitg-fpc32-384-diving48

Video understanding

Generates video. Resolution, clip length, and speed depend on the version and setup.

Source listed: Jun 13, 2025
Parameters: 1.1B
4-bit weights only: ≈ 0.6 GB
License: mit

Compare this model →

Tencent · Downloadable

Hunyuan3D-2.1

3D generation

Creates 3D assets. Check the model card for supported inputs and output formats.

Source listed: Jun 13, 2025
Parameters: Not reported
4-bit weights only: Not reported
License: other

Compare this model →

Meta · Downloadable

OC25

Capabilities unconfirmed

A specialized research model. See the model card for its intended task and supported setup.

Source listed: Jun 13, 2025
Parameters: Not reported
4-bit weights only: Not reported
License: other

Compare this model →

Microsoft · Downloadable

CADFusion

Capabilities unconfirmed

A specialized research model. See the model card for its intended task and supported setup.

Source listed: Jun 13, 2025
Parameters: Not reported
4-bit weights only: Not reported
License: mit

Compare this model →

MiniMax · Downloadable

MiniMax-M1-80k

Text generation

Generates text for conversations and writing.

Source listed: Jun 13, 2025
Parameters: 456.1B
4-bit weights only: ≈ 228 GB
License: apache-2.0

Compare this model →

Google · Downloadable

gemma-3n-E2B-it

VisionVideo understandingTranscription

Understands text and images and responds in text.

Source listed: Jun 12, 2025
Parameters: 5.4B
4-bit weights only: ≈ 2.7 GB
License: gemma

Compare this model →

Google · Downloadable

gemma-3n-E2B

VisionVideo understandingTranscription

Understands text and images and responds in text.

Source listed: Jun 12, 2025
Parameters: 5.4B
4-bit weights only: ≈ 2.7 GB
License: gemma

Compare this model →

StepFun · Downloadable

Step-Audio-AQAA

Capabilities unconfirmed

A specialized research model. See the model card for its intended task and supported setup.

Source listed: Jun 11, 2025
Parameters: 136.7B
4-bit weights only: ≈ 68.3 GB
License: apache-2.0

Compare this model →

OpenAI · Cloud

o3-pro

VisionReasoningTool use

Cloud API for vision, chat & writing, reasoning, tool use. Capabilities are catalog-listed, not independently scored.

Source listed: Jun 10, 2025
License: Provider terms

Compare this model →

M-A-P · Downloadable

340M-20B-GatedDeltaNet-New-hybrid-24-1

Text generation

Generates text, according to its published model architecture (GatedDeltaNetForCausalLM). No task is listed on Hugging Face, so check the model card for intended use.

Source listed: Jun 9, 2025
Parameters: 20B
4-bit weights only: ≈ 10 GB
License: Not disclosed

Compare this model →

OpenBMB · Downloadable

MiniCPM4-0.5B-QAT-Int4-GPTQ-format

Text generation

Generates text for conversations and writing.

Source listed: Jun 9, 2025
Parameters: 500M
4-bit weights only: ≈ 0.3 GB
License: apache-2.0

Compare this model →

M-A-P · Downloadable

340M-20B-GatedDeltaNet-pure-baseline

Text generation

Generates text, according to its published model architecture (GatedDeltaNetForCausalLM). No task is listed on Hugging Face, so check the model card for intended use.

Source listed: Jun 7, 2025
Parameters: 20B
4-bit weights only: ≈ 10 GB
License: Not disclosed

Compare this model →

M-A-P · Downloadable

1.3B-100B-DeltaNet-pure

Text generation

Generates text, according to its published model architecture (DeltaNetForCausalLM). No task is listed on Hugging Face, so check the model card for intended use.

Source listed: Jun 7, 2025
Parameters: 1.3B
4-bit weights only: ≈ 0.7 GB
License: Not disclosed

Compare this model →

M-A-P · Downloadable

1.3B-100B-DeltaNet-hybrid-24-1

Text generation

Generates text, according to its published model architecture (DeltaNetForCausalLM). No task is listed on Hugging Face, so check the model card for intended use.

Source listed: Jun 7, 2025
Parameters: 1.3B
4-bit weights only: ≈ 0.7 GB
License: Not disclosed

Compare this model →

M-A-P · Downloadable

1.3B-100B-DeltaNet-hybrid-12-1

Text generation

Generates text, according to its published model architecture (DeltaNetForCausalLM). No task is listed on Hugging Face, so check the model card for intended use.

Source listed: Jun 7, 2025
Parameters: 1.3B
4-bit weights only: ≈ 0.7 GB
License: Not disclosed

Compare this model →

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.