released version 0.0.2 of the moondream python client today! github.com/vikhyat/moon...
moondream/clients/python at main · vikhyat/moondream
tiny vision language model. Contribute to vikhyat/moondream development by creating an account on GitHub.
github.com
November 22, 2024 at 7:06 AM
released version 0.0.2 of the moondream python client today! github.com/vikhyat/moon...
🌱 #Rhododendron / Aromi #Azalea 'Moondream'
April 24, 2025 at 5:43 PM
🌱 #Rhododendron / Aromi #Azalea 'Moondream'
September 3, 2026 at 5:38 PM
January 10, 2025 at 11:03 AM
The Battle of the Staircase
-rough sketch-
Illustration for Moondream novella
-rough sketch-
Illustration for Moondream novella
January 15, 2026 at 9:19 PM
The Battle of the Staircase
-rough sketch-
Illustration for Moondream novella
-rough sketch-
Illustration for Moondream novella
trying out different captioning models, here's llama joycaption vs moondream v3, im leaning towards joycaption although... it took like 3 mins to caption a single image on my cpu 🥲
October 1, 2025 at 1:36 AM
trying out different captioning models, here's llama joycaption vs moondream v3, im leaning towards joycaption although... it took like 3 mins to caption a single image on my cpu 🥲
Spinning to create a plume of dust. Illustration for my novella Moondream
January 23, 2026 at 9:07 PM
Spinning to create a plume of dust. Illustration for my novella Moondream
just got fired from moondream... my job was to avoid leaking proprietary synthetic data generation techniques in a conference talk.
www.youtube.com/watch?v=T7sx...
www.youtube.com/watch?v=T7sx...
Moondream: how does a tiny vision model slap so hard? — Vikhyat Korrapati
YouTube video by AI Engineer
www.youtube.com
November 19, 2024 at 4:13 AM
just got fired from moondream... my job was to avoid leaking proprietary synthetic data generation techniques in a conference talk.
www.youtube.com/watch?v=T7sx...
www.youtube.com/watch?v=T7sx...
And now I just finished the 2nd draft of my sci-fi story Moondream. 25,731 words. Solidly in "novella" length.
Just finished the rough draft of my gonzo-poetic sci-fi novella, tentatively titled Moondream. Approximately 125 handwritten pages, probably around 22,000 words. Quite short as far as prose books go, but quite long for a guy who usually mostly just draws pictures.
December 11, 2024 at 6:22 PM
And now I just finished the 2nd draft of my sci-fi story Moondream. 25,731 words. Solidly in "novella" length.
November 18, 2024 at 4:55 AM
Moon Duke Fragonard Pours a Glass of Ancient Wine
(Rough sketch for my as-yet unpublished illustrated novella Moondream)
(Rough sketch for my as-yet unpublished illustrated novella Moondream)
January 13, 2026 at 7:32 PM
Moon Duke Fragonard Pours a Glass of Ancient Wine
(Rough sketch for my as-yet unpublished illustrated novella Moondream)
(Rough sketch for my as-yet unpublished illustrated novella Moondream)
Happily doing propaganda for small models and the new generation of open source AI startups in a very nice company (Pleias, Moondream, Prime Intellect). globalventuring.com/corporate/eu...
May 15, 2025 at 2:54 PM
Happily doing propaganda for small models and the new generation of open source AI startups in a very nice company (Pleias, Moondream, Prime Intellect). globalventuring.com/corporate/eu...
Every chapter of my unpublished sci-fi novella Moondream ends with a killer line.
December 5, 2025 at 7:13 PM
Every chapter of my unpublished sci-fi novella Moondream ends with a killer line.
moondream.ai is another
Moondream
Moondream AI - Vision language model for everyone
moondream.ai
July 9, 2025 at 4:49 PM
moondream.ai is another
moondream-2b: a lighting fast vision model moondream.ai
they have a demo, but how do you execute vision models locally?
they have a demo, but how do you execute vision models locally?
Moondream AI
Moondream is a highly efficient open-source vision language model that combines powerful image understanding capabilities with a remarkably small footprint.
moondream.ai
January 9, 2025 at 6:38 PM
moondream-2b: a lighting fast vision model moondream.ai
they have a demo, but how do you execute vision models locally?
they have a demo, but how do you execute vision models locally?
Here’s a great line from my unpublished sci-fi novella Moondream:
December 5, 2025 at 5:26 PM
Here’s a great line from my unpublished sci-fi novella Moondream:
Moondream playground is back, running the latest upcoming version of the model, and now with prompt suggestions! Look at how blazingly fast it is running on an L40S...
Try it out here: moondream.ai/playground
Try it out here: moondream.ai/playground
November 27, 2024 at 4:26 AM
Moondream playground is back, running the latest upcoming version of the model, and now with prompt suggestions! Look at how blazingly fast it is running on an L40S...
Try it out here: moondream.ai/playground
Try it out here: moondream.ai/playground
There have been irritations with the council planning to close our road for a month which has necessitated writing to councils, but I now have this Moondream, which is delicious. Was the overnight prom good? #ConfinedCocktails
August 9, 2025 at 6:02 PM
There have been irritations with the council planning to close our road for a month which has necessitated writing to councils, but I now have this Moondream, which is delicious. Was the overnight prom good? #ConfinedCocktails
Moondream guy wants VLMs to read clocks, I want good multilingual SLMs trained on open data, the frog wants to automate math. Maybe not all realistic but we’re trying.
November 24, 2024 at 7:23 PM
Moondream guy wants VLMs to read clocks, I want good multilingual SLMs trained on open data, the frog wants to automate math. Maybe not all realistic but we’re trying.
Gaze detection will be in the upcoming moondream release!
Live demo: huggingface.co/spaces/moond...
Blog post: moondream.ai/blog/announc...
Live demo: huggingface.co/spaces/moond...
Blog post: moondream.ai/blog/announc...
January 7, 2025 at 1:48 AM
Gaze detection will be in the upcoming moondream release!
Live demo: huggingface.co/spaces/moond...
Blog post: moondream.ai/blog/announc...
Live demo: huggingface.co/spaces/moond...
Blog post: moondream.ai/blog/announc...
Moondream's Photon 2.0: Inference engine for Physical AI
Runs Moondream, Qwen 3.5, and Gemma 4 on NVIDIA H100. Photon outperformed vLLM and SGLang in every throughput test and brought every model online faster, via use of a single 'megakernel' that runs the entire inference on the GPU alone.
Runs Moondream, Qwen 3.5, and Gemma 4 on NVIDIA H100. Photon outperformed vLLM and SGLang in every throughput test and brought every model online faster, via use of a single 'megakernel' that runs the entire inference on the GPU alone.
August 4, 2026 at 4:26 PM
Moondream's Photon 2.0: Inference engine for Physical AI
Runs Moondream, Qwen 3.5, and Gemma 4 on NVIDIA H100. Photon outperformed vLLM and SGLang in every throughput test and brought every model online faster, via use of a single 'megakernel' that runs the entire inference on the GPU alone.
Runs Moondream, Qwen 3.5, and Gemma 4 on NVIDIA H100. Photon outperformed vLLM and SGLang in every throughput test and brought every model online faster, via use of a single 'megakernel' that runs the entire inference on the GPU alone.
Moondream 3
A 9B param, 2B active MoE vision hybrid reasoning vision language model that supports both reasoning and non-reasoning mode. It focus on visually grounded reasoning, where the model references objects and spatial positions in the image while doing said reasoning,
A 9B param, 2B active MoE vision hybrid reasoning vision language model that supports both reasoning and non-reasoning mode. It focus on visually grounded reasoning, where the model references objects and spatial positions in the image while doing said reasoning,
September 21, 2025 at 11:23 PM
Moondream 3
A 9B param, 2B active MoE vision hybrid reasoning vision language model that supports both reasoning and non-reasoning mode. It focus on visually grounded reasoning, where the model references objects and spatial positions in the image while doing said reasoning,
A 9B param, 2B active MoE vision hybrid reasoning vision language model that supports both reasoning and non-reasoning mode. It focus on visually grounded reasoning, where the model references objects and spatial positions in the image while doing said reasoning,
🚀 The GeoAI QGIS Plugin is here 🔥
You can run Moondream vision-language models, object detection, image segmentation (SAM 3), and even train your own geospatial segmentation model end-to-end.
You can run Moondream vision-language models, object detection, image segmentation (SAM 3), and even train your own geospatial segmentation model end-to-end.
December 12, 2025 at 5:48 PM
🚀 The GeoAI QGIS Plugin is here 🔥
You can run Moondream vision-language models, object detection, image segmentation (SAM 3), and even train your own geospatial segmentation model end-to-end.
You can run Moondream vision-language models, object detection, image segmentation (SAM 3), and even train your own geospatial segmentation model end-to-end.
September 8, 2025 at 3:07 AM