正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

#M5StackNew 🎊 The LLM630 Compute Kit is an #AI large language model (#LLM) inference development kit, powered by the #Axera #AX630C SoC with a 3.2 TOPs NPU, it delivers efficient AI inference for tasks like computer vision (CV) and LLM processing.

M5Stack

34,087 subscribers

16,383 次观看 • 1 年前 •via X (Twitter)

科学技术 #M5StackNew #AI #LLM #Axera #AX630C

Anya Rossi• Live Now

Private livecam show

7 条评论

__cdecl 的头像

__cdecl1 年前

后续会出配套的外设配件吗，比如外壳和显示屏，我之前看测试板还能加摄像头

Coral AI News 的头像

Coral AI News1 年前

Coral AI is the most powerful AI for documents. See the difference yourself:

darknessblade 的头像

darknessblade1 年前

Any documentation on how to add it to Home-assistant for a small LLM machine?

Phil Miasup ♨ 的头像

Phil Miasup ♨1 年前

OK

jezza kezza 的头像

jezza kezza1 年前

Got one ordered

Carlo_0001 的头像

Carlo_00011 年前

holy shit :O wow, price?

greenYng 的头像

greenYng1 年前

Have you been able to deploy Phi models, how would the procedure be?

相关视频

#M5StackNew 🎉AI Pyramid & AI Pyramid-Pro Released AI Pyramid & AI Pyramid-Pro are pyramid-shaped high-performance AI PCs purpose-built for local AI inference and edge intelligence deployments. Powered by the #Axera AX8850 SoC, they effortlessly handle complex workloads including visual recognition, multimodal interaction, and on‑device large language model inference, compatible with #Immich (AI local photo album), #Frigate (AI smart security), local Home Assistant and #AIGC deployment. ✨Features ✅Axera AX8850 SoC, 24 TOPS @ INT8 ✅4GB (2+2) High‑Bandwidth Memory (8GB for the pro version): 64‑bit LPDDR4x, 4266 Mbps ✅2× HDMI 2.0 digital A/V output ports(1× output + 1× input for the pro version), 4K@60fps supported ✅Hardware Video Engine: 8K @ 30 fps H.264/H.265 encoding & decoding, 16‑channel 1080p parallel decoding supported ✅Native AXCL Support: One‑click deployment for #CNN、#Transformer、#CLIP、#Whisper、#Llama 3.2、#Qwen3、#InternVL3 and other models; H.264/H.265 simultaneous encode/decode/transcode enabled 🛒 Get Yours: 🔹AI Pyramid: Shopify: AliExpress: 🔹AI Pyramid-Pro: Shopify: AliExpress: #M5Stack #M5StackNew #AI #EdgeAI #AIPC #IoT #Hardware #Makers #SmartHome #Axera #Llama3 #DeepLearning

#M5StackNew 🎉AI Pyramid & AI Pyramid-Pro Released AI Pyramid & AI Pyramid-Pro are pyramid-shaped high-performance AI PCs purpose-built for local AI inference and edge intelligence deployments. Powered by the #Axera AX8850 SoC, they effortlessly handle complex workloads including visual recognition, multimodal interaction, and on‑device large language model inference, compatible with #Immich (AI local photo album), #Frigate (AI smart security), local Home Assistant and #AIGC deployment. ✨Features ✅Axera AX8850 SoC, 24 TOPS @ INT8 ✅4GB (2+2) High‑Bandwidth Memory (8GB for the pro version): 64‑bit LPDDR4x, 4266 Mbps ✅2× HDMI 2.0 digital A/V output ports(1× output + 1× input for the pro version), 4K@60fps supported ✅Hardware Video Engine: 8K @ 30 fps H.264/H.265 encoding & decoding, 16‑channel 1080p parallel decoding supported ✅Native AXCL Support: One‑click deployment for #CNN、#Transformer、#CLIP、#Whisper、#Llama 3.2、#Qwen3、#InternVL3 and other models; H.264/H.265 simultaneous encode/decode/transcode enabled 🛒 Get Yours: 🔹AI Pyramid: Shopify: AliExpress: 🔹AI Pyramid-Pro: Shopify: AliExpress: #M5Stack #M5StackNew #AI #EdgeAI #AIPC #IoT #Hardware #Makers #SmartHome #Axera #Llama3 #DeepLearning

M5Stack

21,084 次观看 • 4 个月前

#M5StackNew 🎉 LLM‑8850 Card — an ultra‑compact M.2 M‑Key 2242 AI accelerator card designed for edge devices It packs the Axera AX8850 SoC delivering 24 TOPS @ INT8 performance into a tiny 42 mm form factor, enabling plug-and-play multimodal large model and video analysis acceleration for host devices such as Raspberry Pi 5 Raspberry Pi , RK3588 Rockchip Electronics , and x86 PCs. ✨Features ✅ High-performance NPU & Native AXCL ✅ High-bandwidth memory & Hardware video engine ✅ Ultra-compact form factor, Rich I/O & Intelligent cooling/power ✅ Secure boot & encryption 🛒 Get Yours: Shopify: #M5Stack #M5StackNew #IoT #Hardware #Makers #EdgeAI #AIAcceleration #LLM #AIoT #M2Card #MachineVision #RaspberryPi #RK3588 #Intel #AMD #Axera #Rockchip

#M5StackNew 🎉 LLM‑8850 Card — an ultra‑compact M.2 M‑Key 2242 AI accelerator card designed for edge devices It packs the Axera AX8850 SoC delivering 24 TOPS @ INT8 performance into a tiny 42 mm form factor, enabling plug-and-play multimodal large model and video analysis acceleration for host devices such as Raspberry Pi 5 Raspberry Pi , RK3588 Rockchip Electronics , and x86 PCs. ✨Features ✅ High-performance NPU & Native AXCL ✅ High-bandwidth memory & Hardware video engine ✅ Ultra-compact form factor, Rich I/O & Intelligent cooling/power ✅ Secure boot & encryption 🛒 Get Yours: Shopify: #M5Stack #M5StackNew #IoT #Hardware #Makers #EdgeAI #AIAcceleration #LLM #AIoT #M2Card #MachineVision #RaspberryPi #RK3588 #Intel #AMD #Axera #Rockchip

M5Stack

12,485 次观看 • 8 个月前

Deterministic inference: getting the exact same output every time you run an LLM with identical inputs. Try it yourself: Powered by EigenAI

Deterministic inference: getting the exact same output every time you run an LLM with identical inputs. Try it yourself: Powered by EigenAI

nader dabit

29,335 次观看 • 7 个月前

Another example of the multiple TTS parallel pipelines pattern. Here's a voice AI agent that speaks both English and Arabic, using a specific model/voice for each language. These are PlayAI voices. The STT, TTS, and LLM inference is all running on Groq Inc. (The LLM is LLama 4 Maverick.)

Another example of the multiple TTS parallel pipelines pattern. Here's a voice AI agent that speaks both English and Arabic, using a specific model/voice for each language. These are PlayAI voices. The STT, TTS, and LLM inference is all running on Groq Inc. (The LLM is LLama 4 Maverick.)

kwindla

13,328 次观看 • 1 年前

This Paper from Google DeepMind is a landmark one. 📚 "Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters" It may have contributed to the 01 Model from OpenAI or the principle may have been long known to OpenAI. The paper basically says - Searching at inference will give you great final result from the LLM. Now from Strawberry model is making use of inference time compute strategies using Search techniques over the response space to improve reasoning. ------ The Podcast is generated with Google's illuminate, the tool trained on AI & science-related Arxiv papers.

This Paper from Google DeepMind is a landmark one. 📚 "Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters" It may have contributed to the 01 Model from OpenAI or the principle may have been long known to OpenAI. The paper basically says - Searching at inference will give you great final result from the LLM. Now from Strawberry model is making use of inference time compute strategies using Search techniques over the response space to improve reasoning. ------ The Podcast is generated with Google's illuminate, the tool trained on AI & science-related Arxiv papers.

Rohan Paul

48,980 次观看 • 1 年前

Multimodal reasoning with Phi-4-reasoning-vision, new work on scaling LLM inference, benchmarking AI agents in network operations, cinematic video generation, adaptive evaluation for LLMs, and using AI to improve individual and population health.

Multimodal reasoning with Phi-4-reasoning-vision, new work on scaling LLM inference, benchmarking AI agents in network operations, cinematic video generation, adaptive evaluation for LLMs, and using AI to improve individual and population health.

Microsoft Research

11,672 次观看 • 3 个月前

Crusoe Managed Inference is now available. Run model inference faster, easier, and more cost-effectively. Large-context AI solutions demand fast outputs, massive throughput, and resilient scaling. Crusoe Managed Inference delivers breakthrough speed & reduced input-token spend.

Crusoe Managed Inference is now available. Run model inference faster, easier, and more cost-effectively. Large-context AI solutions demand fast outputs, massive throughput, and resilient scaling. Crusoe Managed Inference delivers breakthrough speed & reduced input-token spend.

Crusoe

12,096,815 次观看 • 7 个月前

OpenAI's Noam Brown says that while AI model performance scales roughly equivalently with more training or inference compute, the cost of inference is on the order of 100 billion times cheaper

OpenAI's Noam Brown says that while AI model performance scales roughly equivalently with more training or inference compute, the cost of inference is on the order of 100 billion times cheaper

Tsarathustra

250,828 次观看 • 1 年前

Breaking: We can now prove that AI is correct. DeepProve-1 is the first production-ready zkML system to cryptographically verify a full LLM inference. Lagrange has successfully proven the inference of OpenAI’s GPT-2, moving verifiable AI from theory to production: 🧵

Breaking: We can now prove that AI is correct. DeepProve-1 is the first production-ready zkML system to cryptographically verify a full LLM inference. Lagrange has successfully proven the inference of OpenAI’s GPT-2, moving verifiable AI from theory to production: 🧵

LAGRANGE

249,292 次观看 • 10 个月前

🚀BREAKING: Freedom AI is live!🚀 The uncensored LLM you've been waiting for – no limits, no censorship, just raw truth. Powered by our high performance compute infrastructure, it delivers unmatched AI freedom. 💡 Chat about anything 🔒 Privacy-first 🔥 Early access now Join the AI revolution: Retweet and tag 3 friends for a chance to win $NLOV tokens! 🎉

🚀BREAKING: Freedom AI is live!🚀 The uncensored LLM you've been waiting for – no limits, no censorship, just raw truth. Powered by our high performance compute infrastructure, it delivers unmatched AI freedom. 💡 Chat about anything 🔒 Privacy-first 🔥 Early access now Join the AI revolution: Retweet and tag 3 friends for a chance to win $NLOV tokens! 🎉

Neurolov

21,236 次观看 • 1 年前

LLM inference speed with vs. without KV caching: (learn how and why it works below)

LLM inference speed with vs. without KV caching: (learn how and why it works below)

Daily Dose of Data Science

59,218 次观看 • 2 个月前

LLM inference speed with vs. without KV caching: (learn how and why it works below)

LLM inference speed with vs. without KV caching: (learn how and why it works below)

Avi Chawla

395,041 次观看 • 3 个月前

1/ Announcing the first Compute Market on Nockchain: AI inference.

1/ Announcing the first Compute Market on Nockchain: AI inference.

Nockchain

163,655 次观看 • 1 个月前

Our proprietary AI model delivers animated characters on-device in real time, allowing any conversation with LLM-powered chatbots to be easily visualised.

Our proprietary AI model delivers animated characters on-device in real time, allowing any conversation with LLM-powered chatbots to be easily visualised.

Sergey Gonchar

126,369 次观看 • 1 年前

What if your LLM inference automatically got faster the more you used it? Introducing ATLAS from the Together AI Turbo research team. Read more: Here’s Together AI Founder and Chief Scientist Tri Dao introducing ATLAS:

What if your LLM inference automatically got faster the more you used it? Introducing ATLAS from the Together AI Turbo research team. Read more: Here’s Together AI Founder and Chief Scientist Tri Dao introducing ATLAS:

Together AI

143,442 次观看 • 8 个月前

Fuel AI Agent Demo ⛽ Powered by the Fuel Agent Kit. This is just a glimpse of what's possible with the Fuel Agent Kit and our integrations with projects like ai16zdao's Eliza and GOAT (Great Onchain Agent Toolkit).

Fuel AI Agent Demo ⛽ Powered by the Fuel Agent Kit. This is just a glimpse of what's possible with the Fuel Agent Kit and our integrations with projects like ai16zdao's Eliza and GOAT (Great Onchain Agent Toolkit).

Fuel

22,764 次观看 • 1 年前

“Nillion aims to provide privacy to the people who have none.” John Woods CTO of Nillion joins The Defiant to unpack the “blind computer”—a federated network for private storage and compute—with NIL 2.0 weaving tokenomics directly into the tech stack. From private LLM inference to genomics, this is privacy-by-design in action. Listen now👇

“Nillion aims to provide privacy to the people who have none.” John Woods CTO of Nillion joins The Defiant to unpack the “blind computer”—a federated network for private storage and compute—with NIL 2.0 weaving tokenomics directly into the tech stack. From private LLM inference to genomics, this is privacy-by-design in action. Listen now👇

The Defiant

15,503 次观看 • 7 个月前

AI development is increasingly centralised. Gensyn is building an open protocol for machine intelligence - where verifiable AI training, inference, evaluation, and payments can be performed on any device, not just large clusters.

AI development is increasingly centralised. Gensyn is building an open protocol for machine intelligence - where verifiable AI training, inference, evaluation, and payments can be performed on any device, not just large clusters.

gensyn

87,885 次观看 • 6 个月前