Ils ont publié un outil de compression du KV cache, qui fait partie des techniques qu'on a utilisé chez Pruna pour optimiser gpt-oss-120B par ici : replicate.com/prunaai/gpt-...
Leur lib est dispo par ici si tu veux jouer avec : github.com/NVIDIA/kvpress, et ça s'intègre bien dans vLLM
Leur lib est dispo par ici si tu veux jouer avec : github.com/NVIDIA/kvpress, et ça s'intègre bien dans vLLM
prunaai/gpt-oss-120b-fast – Replicate
This is a version of GPT-OSS-120B that is optimised by Pruna AI.
replicate.com
March 18, 2026 at 11:41 AM
Ils ont publié un outil de compression du KV cache, qui fait partie des techniques qu'on a utilisé chez Pruna pour optimiser gpt-oss-120B par ici : replicate.com/prunaai/gpt-...
Leur lib est dispo par ici si tu veux jouer avec : github.com/NVIDIA/kvpress, et ça s'intègre bien dans vLLM
Leur lib est dispo par ici si tu veux jouer avec : github.com/NVIDIA/kvpress, et ça s'intègre bien dans vLLM
January 18, 2026 at 2:29 PM
My friend @aminesaboni.bsky.social (and team) has just released Pruna
github.com/PrunaAI/pruna/
Shoot your questions and remarks :)
(You can as well share it ^^)
github.com/PrunaAI/pruna/
Shoot your questions and remarks :)
(You can as well share it ^^)
GitHub - PrunaAI/pruna: Pruna is a model optimization framework built for developers, enabling you to deliver faster, more efficient models with minimal overhead.
Pruna is a model optimization framework built for developers, enabling you to deliver faster, more efficient models with minimal overhead. - PrunaAI/pruna
github.com
March 17, 2025 at 1:24 PM
My friend @aminesaboni.bsky.social (and team) has just released Pruna
github.com/PrunaAI/pruna/
Shoot your questions and remarks :)
(You can as well share it ^^)
github.com/PrunaAI/pruna/
Shoot your questions and remarks :)
(You can as well share it ^^)
Pruna vient de publier son framework d'optimisation d'inférence (quantization, compilation, caching pour diffusers, etc.). Venez nous mettre une petite étoile, et du feedback sur les méthodes d'optimisation !
github.com/PrunaAI/pruna/
github.com/PrunaAI/pruna/
GitHub - PrunaAI/pruna: Pruna is a model optimization framework built for developers, enabling you to deliver faster, more efficient models with minimal overhead.
Pruna is a model optimization framework built for developers, enabling you to deliver faster, more efficient models with minimal overhead. - PrunaAI/pruna
github.com
March 17, 2025 at 9:57 AM
Pruna vient de publier son framework d'optimisation d'inférence (quantization, compilation, caching pour diffusers, etc.). Venez nous mettre une petite étoile, et du feedback sur les méthodes d'optimisation !
github.com/PrunaAI/pruna/
github.com/PrunaAI/pruna/
Happy to announce the "Awesome AI Efficiency" repo gathering 100+ curated materials on efficient AI
It is designed for a large audience and includes diverse materials.
It is an ongoing project. Do not hesitate to share your feedback/suggestions and star the repo! 🌟
github.com/PrunaAI/awes...
It is designed for a large audience and includes diverse materials.
It is an ongoing project. Do not hesitate to share your feedback/suggestions and star the repo! 🌟
github.com/PrunaAI/awes...
GitHub - PrunaAI/awesome-ai-efficiency: A curated list of materials on AI efficiency
A curated list of materials on AI efficiency. Contribute to PrunaAI/awesome-ai-efficiency development by creating an account on GitHub.
github.com
May 14, 2025 at 1:11 PM
Happy to announce the "Awesome AI Efficiency" repo gathering 100+ curated materials on efficient AI
It is designed for a large audience and includes diverse materials.
It is an ongoing project. Do not hesitate to share your feedback/suggestions and star the repo! 🌟
github.com/PrunaAI/awes...
It is designed for a large audience and includes diverse materials.
It is an ongoing project. Do not hesitate to share your feedback/suggestions and star the repo! 🌟
github.com/PrunaAI/awes...
Our cofounder @bertrand-sharp.bsky.social created this awesome repository, curating the best resources about #AI Efficiency.
Be among the first to⭐️ and contribute!
github.com/PrunaAI/awes...
Be among the first to⭐️ and contribute!
github.com/PrunaAI/awes...
GitHub - PrunaAI/awesome-ai-efficiency: A curated list of materials on AI efficiency
A curated list of materials on AI efficiency. Contribute to PrunaAI/awesome-ai-efficiency development by creating an account on GitHub.
github.com
March 3, 2025 at 2:51 PM
Our cofounder @bertrand-sharp.bsky.social created this awesome repository, curating the best resources about #AI Efficiency.
Be among the first to⭐️ and contribute!
github.com/PrunaAI/awes...
Be among the first to⭐️ and contribute!
github.com/PrunaAI/awes...
Same prompt. Four AI image models. Four completely different results.
Nano Banana 2 Lite, PrunaAI/P-Image-Edit, Ideogram V4 Quality, and GPT-Image-2 each gave the same monster claw + energy drink scene a unique look.
Which one wins? Comment below.
#AI #PromptEngineering
Nano Banana 2 Lite, PrunaAI/P-Image-Edit, Ideogram V4 Quality, and GPT-Image-2 each gave the same monster claw + energy drink scene a unique look.
Which one wins? Comment below.
#AI #PromptEngineering
July 24, 2026 at 12:07 PM
Same prompt. Four AI image models. Four completely different results.
Nano Banana 2 Lite, PrunaAI/P-Image-Edit, Ideogram V4 Quality, and GPT-Image-2 each gave the same monster claw + energy drink scene a unique look.
Which one wins? Comment below.
#AI #PromptEngineering
Nano Banana 2 Lite, PrunaAI/P-Image-Edit, Ideogram V4 Quality, and GPT-Image-2 each gave the same monster claw + energy drink scene a unique look.
Which one wins? Comment below.
#AI #PromptEngineering
August 4, 2025 at 3:19 PM
i like rotating them around but current roster is hitting replicate.com api with:
prunaai/hidream-l1-fast
bytedance/seedream-4.5
black-forest-labs/flux-2-max
openai/gpt-image-2
im a huge fan of seedream rn. I've never really done loras much, i like getting the image models as "base" as i can
prunaai/hidream-l1-fast
bytedance/seedream-4.5
black-forest-labs/flux-2-max
openai/gpt-image-2
im a huge fan of seedream rn. I've never really done loras much, i like getting the image models as "base" as i can
July 13, 2026 at 4:15 PM
i like rotating them around but current roster is hitting replicate.com api with:
prunaai/hidream-l1-fast
bytedance/seedream-4.5
black-forest-labs/flux-2-max
openai/gpt-image-2
im a huge fan of seedream rn. I've never really done loras much, i like getting the image models as "base" as i can
prunaai/hidream-l1-fast
bytedance/seedream-4.5
black-forest-labs/flux-2-max
openai/gpt-image-2
im a huge fan of seedream rn. I've never really done loras much, i like getting the image models as "base" as i can
Replicate and @PrunaAI join forces to bring Wan2.2 to life!
A major upgrade in cinematic quality, smoother movements, and instruction following
replicate.com/wan-video/w...
A major upgrade in cinematic quality, smoother movements, and instruction following
replicate.com/wan-video/w...
July 29, 2025 at 8:01 PM
Replicate and @PrunaAI join forces to bring Wan2.2 to life!
A major upgrade in cinematic quality, smoother movements, and instruction following
replicate.com/wan-video/w...
A major upgrade in cinematic quality, smoother movements, and instruction following
replicate.com/wan-video/w...
Il y en a quelques de déjà compressés sur notre repo HF (comme celui ci par ex: huggingface.co/PrunaAI/Phi-...), d'autres arriveront bientôt, notamment parmi les modèles récents populaires.
PrunaAI/Phi-3-mini-128k-instruct-GGUF-Imatrix-smashed · Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
January 29, 2025 at 5:18 PM
Il y en a quelques de déjà compressés sur notre repo HF (comme celui ci par ex: huggingface.co/PrunaAI/Phi-...), d'autres arriveront bientôt, notamment parmi les modèles récents populaires.
Pruna offers solutions to reduce AI model warmup times in PyTorch. By using efficient warmup elimination techniques, developers can accelerate deployment & LoRA switching. Insights: https://huggingface.co/blog/PrunaAI/torch-compile-zero-warmup
January 29, 2026 at 11:00 AM
Pruna offers solutions to reduce AI model warmup times in PyTorch. By using efficient warmup elimination techniques, developers can accelerate deployment & LoRA switching. Insights: https://huggingface.co/blog/PrunaAI/torch-compile-zero-warmup
You want to get started?
- Check the documentation: docs.pruna.ai/en/latest/
- Check the Youtube tutorials: www.youtube.com/@PrunaAI
We hope that you will like it and would love to hear your thoughts!
- Check the documentation: docs.pruna.ai/en/latest/
- Check the Youtube tutorials: www.youtube.com/@PrunaAI
We hope that you will like it and would love to hear your thoughts!
Welcome to Pruna’s documentation! | Pruna documentation
docs.pruna.ai
December 12, 2024 at 1:10 PM
You want to get started?
- Check the documentation: docs.pruna.ai/en/latest/
- Check the Youtube tutorials: www.youtube.com/@PrunaAI
We hope that you will like it and would love to hear your thoughts!
- Check the documentation: docs.pruna.ai/en/latest/
- Check the Youtube tutorials: www.youtube.com/@PrunaAI
We hope that you will like it and would love to hear your thoughts!
Join our next event with Nils Fleischmann (Pruna AI) about optimizing the ever-growing AI models!
𝗦𝗮𝘃𝗲 𝘁𝗵𝗲 𝗱𝗮𝘁𝗲! June 17th, 7pm (CET)
𝗝𝗼𝗶𝗻: discord.com/invite/bCuya...
#LLM #PrunaAI #AGI #AI #MunichNLP
𝗦𝗮𝘃𝗲 𝘁𝗵𝗲 𝗱𝗮𝘁𝗲! June 17th, 7pm (CET)
𝗝𝗼𝗶𝗻: discord.com/invite/bCuya...
#LLM #PrunaAI #AGI #AI #MunichNLP
Join the Munich🥨NLP Discord Server!
We <3 NLP | 1162 members
discord.com
May 19, 2025 at 6:13 PM
Join our next event with Nils Fleischmann (Pruna AI) about optimizing the ever-growing AI models!
𝗦𝗮𝘃𝗲 𝘁𝗵𝗲 𝗱𝗮𝘁𝗲! June 17th, 7pm (CET)
𝗝𝗼𝗶𝗻: discord.com/invite/bCuya...
#LLM #PrunaAI #AGI #AI #MunichNLP
𝗦𝗮𝘃𝗲 𝘁𝗵𝗲 𝗱𝗮𝘁𝗲! June 17th, 7pm (CET)
𝗝𝗼𝗶𝗻: discord.com/invite/bCuya...
#LLM #PrunaAI #AGI #AI #MunichNLP
#GenAI, behind the scenes: concrete, metals, heat and vapor
I've been invited by Pruna AI to talk about the heavy material and infrastructures reality behind GenAI services. ⬇️
🗓️ April 15th 2025
📍 app.livestorm.co/pruna-ai/gen...
🙏 Hosted by @PrunaAI
I've been invited by Pruna AI to talk about the heavy material and infrastructures reality behind GenAI services. ⬇️
🗓️ April 15th 2025
📍 app.livestorm.co/pruna-ai/gen...
🙏 Hosted by @PrunaAI
Gen AI, behind the scenes: concrete, metals, heat and vapor | Pruna AI
Debates regarding gen AI are often performed in the positive vs negative environmental impacts axis. If negative impacts assessment is still a recent yet buzzling research field, positive impacts a...
app.livestorm.co
March 24, 2025 at 6:38 PM
#GenAI, behind the scenes: concrete, metals, heat and vapor
I've been invited by Pruna AI to talk about the heavy material and infrastructures reality behind GenAI services. ⬇️
🗓️ April 15th 2025
📍 app.livestorm.co/pruna-ai/gen...
🙏 Hosted by @PrunaAI
I've been invited by Pruna AI to talk about the heavy material and infrastructures reality behind GenAI services. ⬇️
🗓️ April 15th 2025
📍 app.livestorm.co/pruna-ai/gen...
🙏 Hosted by @PrunaAI
PVideo Avatar — генератор видео с ИИ-аватарами Загружаешь
фото и аудио (или описываешь голосом текстом) — получаешь видео с говорящим аватаром. Поддерживает русский язык. Работает через платформу Replicate, где можно протестировать модель.
Проверить: https://replicate.com/prunaai/p-video-avatar
фото и аудио (или описываешь голосом текстом) — получаешь видео с говорящим аватаром. Поддерживает русский язык. Работает через платформу Replicate, где можно протестировать модель.
Проверить: https://replicate.com/prunaai/p-video-avatar
August 18, 2026 at 11:29 AM
PVideo Avatar — генератор видео с ИИ-аватарами Загружаешь
фото и аудио (или описываешь голосом текстом) — получаешь видео с говорящим аватаром. Поддерживает русский язык. Работает через платформу Replicate, где можно протестировать модель.
Проверить: https://replicate.com/prunaai/p-video-avatar
фото и аудио (или описываешь голосом текстом) — получаешь видео с говорящим аватаром. Поддерживает русский язык. Работает через платформу Replicate, где можно протестировать модель.
Проверить: https://replicate.com/prunaai/p-video-avatar
Two seconds per image is seriously fast. P-Image is now on Magnific with 13 aspect ratios and 1K or 2K output. Change one detail, see it, adjust again while the idea is still fresh.
x.com/magnific/st...
x.com/magnific/st...
Magnific (@magnific) on X
P-Image by @ideogram_ai x @PrunaAI is now on Magnific
→ Images in just 2 seconds
→ 13 different aspect ratios
→ 1K or 2K resolution
Available now on Magnific
x.com
August 5, 2026 at 6:41 PM
Two seconds per image is seriously fast. P-Image is now on Magnific with 13 aspect ratios and 1K or 2K output. Change one detail, see it, adjust again while the idea is still fresh.
x.com/magnific/st...
x.com/magnific/st...
September 23, 2025 at 4:40 PM
Pruna AI’s Open-Source Framework Boosts AI Efficiency in 2025
wiobs.com/pruna-ais-op...
#PrunaAI #AIEfficiency #OpenSourceAI #SustainableTech #AICompression #ImageGeneration #VideoGeneration #DeveloperTools #TechTrends2025 #GreenAI
wiobs.com/pruna-ais-op...
#PrunaAI #AIEfficiency #OpenSourceAI #SustainableTech #AICompression #ImageGeneration #VideoGeneration #DeveloperTools #TechTrends2025 #GreenAI
Pruna AI’s Open-Source Framework Boosts AI Efficiency in 2025 - WIOBS
Pruna AI’s new open-source framework enhances AI efficiency with compression techniques, targeting image and video...
wiobs.com
March 20, 2025 at 10:46 AM
Pruna AI’s Open-Source Framework Boosts AI Efficiency in 2025
wiobs.com/pruna-ais-op...
#PrunaAI #AIEfficiency #OpenSourceAI #SustainableTech #AICompression #ImageGeneration #VideoGeneration #DeveloperTools #TechTrends2025 #GreenAI
wiobs.com/pruna-ais-op...
#PrunaAI #AIEfficiency #OpenSourceAI #SustainableTech #AICompression #ImageGeneration #VideoGeneration #DeveloperTools #TechTrends2025 #GreenAI
Thank you for contributing to Pruna OSS!
⭐️ Star our OSS: buff.ly/Y9j29GO
📚 API & quickstart: buff.ly/VLgjLgw
🌐 Website: www.pruna.ai
⭐️ Star our OSS: buff.ly/Y9j29GO
📚 API & quickstart: buff.ly/VLgjLgw
🌐 Website: www.pruna.ai
GitHub - PrunaAI/pruna: Pruna is a model optimization framework built for developers, enabling you to deliver faster, more efficient models with minimal overhead.
Pruna is a model optimization framework built for developers, enabling you to deliver faster, more efficient models with minimal overhead. - PrunaAI/pruna
buff.ly
June 9, 2026 at 3:03 PM
Thank you for contributing to Pruna OSS!
⭐️ Star our OSS: buff.ly/Y9j29GO
📚 API & quickstart: buff.ly/VLgjLgw
🌐 Website: www.pruna.ai
⭐️ Star our OSS: buff.ly/Y9j29GO
📚 API & quickstart: buff.ly/VLgjLgw
🌐 Website: www.pruna.ai