- 135
- 56 423
Second State
United States
เข้าร่วมเมื่อ 18 ก.ค. 2019
Fast, safe, portable & serverless. Deploy Rust functions in edge computing, Jamstack, SaaS and service mesh applications.
WasmEdge WebAssembly runtime has been hosted as a sandbox project by CNCF under the Linux Foundation. One of the fastest WebAssembly runtimes on the market today.
GitHub github.com/WasmEdge
Documentation wasmedge.org/docs/
WasmEdge WebAssembly runtime has been hosted as a sandbox project by CNCF under the Linux Foundation. One of the fastest WebAssembly runtimes on the market today.
GitHub github.com/WasmEdge
Documentation wasmedge.org/docs/
CPU vs GPU: A whispter API server running on WasmEdge and Docker with WebGPU
CPU ≈ 22 seconds
GPU ≈ 2.2 seconds
github.com/LlamaEdge/LlamaEdge/blob/main/docker/webgpu.md
GPU ≈ 2.2 seconds
github.com/LlamaEdge/LlamaEdge/blob/main/docker/webgpu.md
มุมมอง: 90
วีดีโอ
WasmEdge Community Meeting #31 LLVM JIT support, Wasm GC Proposal with Kotlin, Run LLM on Windows OS
มุมมอง 7221 วันที่ผ่านมา
The team discussed the features and updates of the recently released 0.14 update, including the Git feature, new compiler, and language support. They also explored the performance improvement of compiling and branding a Western file on the fly, the differences between Ablet and Jed versions, and the strategy to improve the development experience with Kotlin. Lastly, they discussed the new proce...
Local LLM with LlamaEdge: Phi-3 reads Nvidia quarterly financial reports
มุมมอง 162หลายเดือนก่อน
What did today’s @NVIDIA financial statement say? Shall I buy $NVDA? I started a Phi-3 LLM, released by @Microsoft yesterday, on my own computer and asked it. DIY: github.com/GaiaNet-AI/node-configs/blob/main/phi-3-medium-instruct-4k/README.md Tech stack: github.com/LlamaEdge/LlamaEdge Try my node: 0xb81f880e9bd28dbddcfd56bd3b9df0390b409b52.gaianet.network
WasmEdge Community Meeting #30 LLM RAG app
มุมมอง 76หลายเดือนก่อน
1.WASMEDGE 0.14.0 PREVIEW WasmEdge New features @hydai - Maintainer of WasmEdge 2.LFX MENTORSHIP OSPP LFX mentorship term 2 and OSPp 2024 araround the corner. Learn the opportunitiesto become Contributor to WasmEdge @juntao - Maintainer 3.BUILD A RAG APPLICATION WITH ONE SINGLE WASM BINARYBuild a complete RAG LLM serviceusing personal knowledge base @apepkuss- LlamaEdge
WasmEdge Community Meeting #29 Llama + Lima + WASI-NN RPC demo, YOLO, Run LLMs with Docker on GPU
มุมมอง 783 หลายเดือนก่อน
Join our next community meeting next month👉docs.google.com/document/d/1iFlVl7R97Lze4RDykzElJGDjjWYDlkI8Rhf8g4dQ5Rk/edit In the realm of Kubernetes and AI, the WasmEdge Community Meeting #29 showcased innovative approaches to deploying and managing AI applications. LLAMA LIMA WASI-NNRPC Demo The first segment, presented by Akihiro Suda, introduced a groundbreaking demo involving LLAMA, LIMA, and...
Google Summer of Code 2024 PROJECTS IDEAS
มุมมอง 783 หลายเดือนก่อน
Boost your resume! Check out the details of WasmEdge's GSoC 2024 project ideas. If you have any questions, comment below these issues👉 github.com/WasmEdge/WasmEdge/issues?q=is:issue is:open label:GSoC @hydai WasmEdge Maintainer
Wasm Functions on Civo Cloud by Saiyam
มุมมอง 583 หลายเดือนก่อน
Demonstrating Wasm functions written in Rust on Civo Talos Kubernetes via the WasmEdge shim. The presentation covers insights into Talos and Wasm extensions. @SaiyamPathak Field CTO @Civo cloud, CNCF Ambassor
Demo: Run on Mac multi-modal open source LLM Llava + Google's Gemma-WasmEdge Community Meeting March
มุมมอง 1023 หลายเดือนก่อน
Llava Demo where you'll learn how to chat with your images locally - a showcase by DM4 a dedicated WasmEdge Committer. Interact with large multi-modal models on your own computer! Ask it what’s in the picture! Complete tutorial: www.secondstate.io/articles/llava-v1.6-vicuna-7b/
WasmEdge Community Meeting #28 Llava Demo, WasmEdge on Civo Talos K8s, GSoC
มุมมอง 933 หลายเดือนก่อน
Join us in the exciting 28th edition of the WasmEdge Community Meeting! This video is packed with insightful presentations and discussions that any WebAssembly enthusiast wouldn't want to miss. 🛠️ First up, we have an intriguing Llava Demo where you'll learn how to chat with your images locally - a showcase by a dedicated WasmEdge Committer. 💻 Next, we delve into the world of Wasm functions on ...
Create an LLM web service with LlamaEdge on a MacBook, deploy it on a NVIDIA device.
มุมมอง 2113 หลายเดือนก่อน
With LlamaEdge, you can develop and compile LLM apps on one machine, and run the apps on another machine. Learn more: llamaedge.com/ Source code: github.com/LlamaEdge/LlamaEdge/
Orchestrate and move an LLM app across CPUs, GPUs and NPUs using LlamaEdge
มุมมอง 714 หลายเดือนก่อน
LlamaEdge application can run without change across devices with different Operation Systems, CPUs and GPUs. The portable workload can be managed and orchestrated by container tools like Kubernetes in a heterogeneous cluster of machines. Learn more: llamaedge.com/ Source code: github.com/LlamaEdge/LlamaEdge/
Demo: Run Llava, the multimodal LLM across Devices and ask it about pictures
มุมมอง 1234 หลายเดือนก่อน
Run Llava: the multimodal LLM on WasmEdge! Portable 2MB binary that runs on any CPU or GPU device. The video shows it running on a Nvidia server. You can run it on your MacBook without any change. Rust source code for the app: github.com/second-state/WasmEdge-WASINN-examples/tree/master/wasmedge-ggml/llava
WASI Preview 2 Discussion - Bailey Hayes (WasmEdge community meeting Feb)
มุมมอง 3064 หลายเดือนก่อน
WASI Preview 2 is a milestone for the Wasm community. It includes wasi-io, wasi-clocks, wasi-random, wasi-filesystem, wasi-sockets, wasi-cli, and wasi-http APIs. PPT slides: docs.google.com/presentation/d/1fln7vdw3IFZS508WWa-i0qD2rKLnzHCwQw51LiUe2Wg/edit#slide=id.p1 Bailey's GitHub Handle: @ricochet WasmCloud Maintainer, Director on the Bytecode Alliance Technical Steering Committee
LFX Mentorship 01-Mar-May term 2024 Introduction -Hydai
มุมมอง 654 หลายเดือนก่อน
WasmEdge's LFX Mentorship 01-Mar-May term 2024 projects are on LLMs
Mewz Introduction - Saza & AiNozaki #wasm #wasi #webassembly
มุมมอง 904 หลายเดือนก่อน
github.com/Mewz-project/Mewz docs.google.com/presentation/d/1L5lcJ8aR41GM1wCX3k9FBq1zF4HxkKb6s7yvdPLu2RA/edit?usp=sharing
WasmEdge Community Meeting #27 Mewz, WASI Preview 2, LFX Mentorship Spring 2024
มุมมอง 1404 หลายเดือนก่อน
WasmEdge Community Meeting #27 Mewz, WASI Preview 2, LFX Mentorship Spring 2024
Self-host StableLM-2-Zephyr-1.6B. Portable across GPUs CPUs OSes
มุมมอง 1515 หลายเดือนก่อน
Self-host StableLM-2-Zephyr-1.6B. Portable across GPUs CPUs OSes
Run Nous-Hermes-2-Mixtral-8x7B on Mac + Across Devices with 1 single command
มุมมอง 2965 หลายเดือนก่อน
Run Nous-Hermes-2-Mixtral-8x7B on Mac Across Devices with 1 single command
Self host Mixtral-8x7B MoE LLM on Mac+cross devices, 2MB AI inference app fully portable
มุมมอง 3375 หลายเดือนก่อน
Self host Mixtral-8x7B MoE LLM on Mac cross devices, 2MB AI inference app fully portable
Mistral-7B-Instruct-v0.2 on Macbook M2. Open source LLM solves math problem
มุมมอง 3935 หลายเดือนก่อน
Mistral-7B-Instruct-v0.2 on Macbook M2. Open source LLM solves math problem
Demo: Mixtral 8x7B on Jetson Orin. Run any open source llms on Huggging Face with LlamaEdge
มุมมอง 1215 หลายเดือนก่อน
Demo: Mixtral 8x7B on Jetson Orin. Run any open source llms on Huggging Face with LlamaEdge
WasmEdge Community Meeting #26 Run Mixtral on Mac demo; WasmEdge LLM support; Wasm@LangChain
มุมมอง 1506 หลายเดือนก่อน
WasmEdge Community Meeting #26 Run Mixtral on Mac demo; WasmEdge LLM support; Wasm@LangChain
Easy Setup! Self-host Mixtral-8x7B across devices with a 2M inference app
มุมมอง 2086 หลายเดือนก่อน
Easy Setup! Self-host Mixtral-8x7B across devices with a 2M inference app
WasmEdge Community Meeting #25: Use docker to run WasmEdge and LLMs
มุมมอง 1166 หลายเดือนก่อน
WasmEdge Community Meeting #25: Use docker to run WasmEdge and LLMs
Run Berkeley's Starling 7B LLM across devices without Python hassles
มุมมอง 3627 หลายเดือนก่อน
Run Berkeley's Starling 7B LLM across devices without Python hassles
Run LLM Microsoft's Orca-2-13B across platforms! Easy setup with Rust+Wasm, no heavy Python needed
มุมมอง 2367 หลายเดือนก่อน
Run LLM Microsoft's Orca-2-13B across platforms! Easy setup with Rust Wasm, no heavy Python needed
Run WizardCoder Python 7B on your devices
มุมมอง 1377 หลายเดือนก่อน
Run WizardCoder Python 7B on your devices
Run Mistral-7b-instruct-v0.1 LLM on Your Device | Setup & Works as OpenAI API Server!
มุมมอง 3707 หลายเดือนก่อน
Run Mistral-7b-instruct-v0.1 LLM on Your Device | Setup & Works as OpenAI API Server!
Run dolphin-2.2-yi-34b on IoT Devices (Also works as a Private OpenAI API Server)
มุมมอง 2817 หลายเดือนก่อน
Run dolphin-2.2-yi-34b on IoT Devices (Also works as a Private OpenAI API Server)
What type of M2?
M2 chip with 16G mem
@@SecondStateInc How do you think it works with various AI-tasks? I'm thinking of buying one.
@@ToreSG705 7B models work on Intel chips too, just slower
Amazing job guys. I believe you are hitting something.
WTF MEWZ IS STUPIDLY AWESOME
We got any slides for Mewz talk?
docs.google.com/presentation/d/1L5lcJ8aR41GM1wCX3k9FBq1zF4HxkKb6s7yvdPLu2RA/edit#slide=id.g28ec80b4925_0_328
Are there any plans to make API protocol compatible with ollama so we could run open-interpreter and crewai with wasmedge stack? Is there any more appropriate channel for technical communication?
Please come join our discord. link is not allowed in comment. check out @secondstateinc or @realwasmedge twitter bio for the discord invite
I do not believe we need Ollama for open-interpreter or crewai. They work with OpenAI APIs, which we support.
or you can also go to our github to raise an issue. I have attached an article in the video description!
Does the WASM model store the whole model checkpoint within his memory space, and if yes, how can a 7B model fit the 4 GB maximum WASM linear memory? How does the WASM module use the CUDA, and in this case, the Apple Metal APIs? Great video!
Any insight?
That's impressive. Congrats! Would it be possible to use WasmEdge+Mixtral as a replacement for LMStudio and run the open interpreter with the command 'interpreter --local'? So, instead of LLM Studio, it would use your implementation. That would be something awesome.
Yes, absolutely! This article may be helpful www.secondstate.io/articles/mixtral-8-7b/. Please let us know if you have any questions.
good job！
Keep up good work! 👍
Sorry man. This explains nothing.
www.secondstate.io/articles/wizardcoder-python-7b/ in the description
i dont consider 18GB vram useage for iot 🙂
Cool
Promo*SM
And I thought my .diff summarizer was high speed 😂 well done
I think wasm on riscv is the future. The earlier the work for this starts the better.
The future is now
awesome
Many thanks to those on your team willing to present in their non-native language so that others from the English speaking sim mimicry may benefit from this very exciting work.
Too fast.
Notesd and thanks for pointing this out. You could aslo check out our docs. wasmedge.org/book/en/frameworks/serverless/secondstate.html
You could not have chosen a smaller font size.
Thanks for pointing this out. We will pay attention next time. You could also check out the text version of this guideline. wasmedge.org/book/en/os/sel4.html
@@SecondStateInc Thank you.
rust has a briiliant future!
Hello my friend, I try to use that but when I configure the devchain as provider is unreachable :-( why thanks Fabrizio
V8 comes with a virtual machine for web assembly built in though..
V8 has a JS engine bundled with its WASM engine. It sort of defeats the whole purpose of “lightweight”. It is also primarily optimized for the browser use case - meaning it has no AOT or other server specific extensions.
Hi amazing video. I plan to launch my own coin but still confused about total supply vs liquidity on pancake, allocation of pool keeping or not the tokens etc. There anyway if you a moment I can dm you about it ;-) really appreciate it mate
Nice!
Any practical example? Any git repo please!
Hi, Please check out the CNCF WasmEdge project github.com/WasmEdge/WasmEdge
Amazing Information !
is it same with uniswap org?the oasis eth uniswap
So this is private, no front running, and possibly lower fee?
Um, this is a GAME CHANGER!
Love that hot dog recognition =)) Great Silicon Valley reference.
Thanks!
I know I am kind of randomly asking but do anybody know of a good site to stream new movies online ?
Why china?
Great stuff
Really interesting tech. I was searching for serverless rust and came across secondstate.io. Question: How do I setup serverless environment on say Digital Ocean, using the information from the article deno-webassembly-rust-wasi? Thanks!
If you just a few functions, you can deploy them as a Nodejs app github.com/second-state/ssvm-nodejs-starter The serverless FaaS platform you can deploy yourself github.com/second-state/wasm-joey Or just use the Second State Functions www.secondstate.io/faas/
Is the smart contact code from ETH compatible to ETC? Can I use the same Smart Contract code?
Yes, you can!
ssvm nb!!
Not sure, why this video has to run at x3
Hey Tim, Glad I found your channel. Excited to check out this video and your other one on hi-perf Rust/JS/wasm! Are you on Twitter?
Yes. you can find Tim via @mistermac2008
@@SecondStateInc Awesome, followed
wow nice. I'm installing it right now. Thank you.
You could also try our FaaS for free if you do not want to run your own nodejs servers. :) Here is a tutorial: www.secondstate.io/articles/faas-image-classification/
Nice! Love the videos - do you have a LBRY or Odysee channel yet ?
We have a Discord channel. Welcome to join! discord.com/invite/MjxtURG
I feel over time the cost of using it will increase a lot. There is an alternative to it stormyapp.com
Fantastic work
Will there be support added to also give an option for open in VS Codespaces?
Juntá lo com o Chrome OS, será perfeito para mim. Obrigado por compartilhar, agora é aguardar chegar. Parabéns pelo vídeo foi bem explicativo.
What should we do if we cant compile on the first step
This will be really nice for working on the go! I've sometimes found myself trying to set an environment up on someone else's computer to work for a couple hours, and I feel bad for installing stuff on their computer. This will be great for that situation (and many others)!
My face at 1:46 😱 this is a revolutionary step for web development
So basically gitpod.io built into Github? Perfect!