Fireworks AI

?? Announcing FireOptimizer/Multi-LoRA ?? I didn't expect what I considered to be a small feature launched last year delivered a powerful impact to our customers. I'm excited to announce Multi-LoRA, an important component of FireOptimizer. Personalized experiences are critical to driving greater usage, retention and customer satisfaction for your product. Without Multi-LoRA, deploying hundreds of fine-tuned models on separate GPUs would be prohibitively expensive. With Multi-LoRA, you can now deliver personalized experiences across thousands of users and use cases, without scaling your costs! More specifically, Multi-LoRA has benefits below: -- Fine-tune and serve hundreds of personalized LoRA models at the same cost as a single base model, which is just $0.2/1M tokens for Llama3.1 8B -- 100x cost-efficiency compared to serving 100 fine-tuned models without Multi-LoRA on other platforms with per-GPU pricing -- Convenient deployment on Fireworks Serverless with per-token pricing and competitive inference speeds, or Fireworks On-Demand and Reserved for larger workloads Multi-LoRA is part of FireOptimizer, our adaptation engine designed to customize and enhance AI model performance for your unique use cases and workload. FireOptimizer capabilities include Adaptive Speculative Execution (https://lnkd.in/ejdD-wGG), that enables up to 3x latency improvements, Customizable Quantization (https://lnkd.in/dwpTU233), to precisely balance speed and quality, and LoRA Fine-Tuning (https://lnkd.in/et2UFzDy) to customize and improve model performance. ?Cresta uses Multi-LoRA to personalize their Knowledge Assist feature for each individual customer on the Fireworks enterprise platform. "Fireworks' Multi-LoRA capabilities align with Cresta's strategy to deploy custom AI through fine-tuning cutting-edge base models. It helps unleash the potential of AI on private enterprise data." - Tim Shi, Co-Founder and CTO of Cresta ?Brainiac Labs helps businesses leverage their proprietary data to fine-tune and deploy models using Multi-LoRA on the Fireworks self-serve platform. “Using Fireworks, clients with limited AI expertise can successfully maintain and improve the solutions I provide. Additionally, students in my course are able to complete real-world fine-tuning projects, dedicating just a few hours per week to the process.” - Scott Kramer, CEO of Brainiac Labs ?? Read more in our blog post https://lnkd.in/d3_HGRqy

?? Now on Fireworks: The new Qwen QwQ model focuses on advancing AI reasoning, and showcases the power of open models to match closed frontier model performance. ?? QwQ-32B-Preview is an experimental release, comparable to o1 and surpassing GPT-4o and Claude 3.5 Sonnet on analytical and reasoning abilities across GPQA, AIME, MATH-500 and LiveCodeBench benchmarks. Fireworks hosts QwQ-32B-Preview on Serverless, where it’s available immediately for fast inference, paid-per-token with no cold boots. This model is served experimentally, so be aware that Fireworks may undeploy the model with 2 weeks notice. Fireworks also hosts QwQ-32B-Preview on On-Demand. On-Demand lets you deploy these models with 1 line of code and use them on private, scalable GPUs powered by Fireworks’ blazing fast and hyper-efficient serving engine. QwQ on Fireworks Playground: https://lnkd.in/gQ-KACMw Get Started with Fireworks: https://lnkd.in/gx8yxM6i

Super excited to moderate this panel of AI experts at #AWSreInvent next week. You'll hear from Marwan Sarieddine of Anyscale, Pradeep Prabhakaran of Cohere and Pranay Bhatia of Fireworks AI, on what they recommend to "Build you AI Stack" based on lessons learned from launching AI applications with a variety of customers. Meet us at the Bollinger room at the Wynn hotel, Wed Dec 4 1:00 pm.! And checkout MongoDB ‘s schedule at #AWSreInvent! ?? https://lnkd.in/gJ9YkYUr

Want a chance to win a Macbook M4 Pro? We're teaming up with LlamaIndex, Fireworks AI, and Weaviate to gather insights into how companies are building and deploying AI — and we need your help. Fill our 4-minute anonymous survey and: 1. Get early access to industry insights 2. Enter to win a MacBook M4 Pro ?? The survey is open to anyone involved in the AI development process — from developers and engineers to product teams and executives. About the Survey We want to learn from your experience—the tools you trust, the challenges you face, and the strategies that work. Here's what the survey covers: - AI Development Journey - Team & Technology - Challenges & Evaluation - Production Use Cases - Impact & Plans The results will be published in January 2025, but you'll get early access as a thank-you for sharing your insights. Fill out the survey here:

?? How Upwork and Fireworks AI Deliver Faster, Smarter Proposals for Freelancers Crafting the perfect proposal can be a challenge for freelancers. But Upwork's new proposal writer feature, now powered by Fireworks, is making it easier for freelancers to pitch their skills effectively. Here’s what makes it stand out: ? Real-time proposal drafts tailored to a freelancer's skills and client needs. ? Ultra-fast AI inference for seamless interactions, powered by Fireworks' FireAttention v2 technology. ? Custom Llama-3.1 LoRA models that enhance content relevance and accuracy. ? Scalable performance to serve millions of freelancers globally. What this means for Upwork's freelancer community: ?? Freelancers save time and effort with instant, personalized proposals. ?? Clients receive better-matched pitches, improving marketplace efficiency. Learn more in our blog post: https://lnkd.in/dhBPU3Mp #upwork #fireworksai #llmsinproduction #genai #genaisuccess #llmops #finetuning #llama3

?? Fireworks is thrilled to partner with our friends at MongoDB for Amazon Web Services (AWS) re:Invent 2024 in Las Vegas! Join us from December 2-6 to explore how we're driving innovation in AI and empowering developers to build smarter, faster applications. ?? Find us at Booth #1406, inside the MongoDB partner enclosure ?? What’s happening: ?? Live demos showcasing how Fireworks accelerates AI development ?? Expert insights on deploying production-ready AI with MongoDB Atlas + Fireworks ?? Exclusive looks at groundbreaking solutions built for builders like you ?? We can’t wait to connect and show you what’s possible with Fireworks and MongoDB. Stop by and say hi to the team Lin Qiao, Sid Rabindran, Bardia Shahali, Alan Hsia! ?? We can't wait to see you there! Details here: https://lnkd.in/eARmv4ta #Mongodb #Fireworksai #genai #llms #DevelopersUnite #AWSreInvent #reinvent2024

?? Introducing Fireworks f1 ?? A compound AI model specialized in complex reasoning. f1 is the first reasoning system over open models to beat GPT-4o and Claude 3.5 Sonnet across hard coding, chat and math benchmarks. At Fireworks AI, we believe the future of AI is shifting to compound AI systems that combine specialized models and tools to achieve better performance, reliability and control. However building compound AI systems is difficult and time-consuming, so we set out to fix that. Today, we’re releasing a first step in that direction. f1 is a compound AI model specialized in complex reasoning, that interweaves multiple open models at the inference layer. f1 enables developers to access the power of compound AI with the simplicity of prompting. Using prompt as the universal declarative programming language for Gen AI application building, developers can describe what they want to achieve without needing to specify exactly how to accomplish it. ? two variants now available in preview: f1 and f1-mini ? access the preview on Fireworks AI Playground for free ? get on to waitlist for free early access to the f1 API We invite you to help us improve these models and shape the future of compound AI. Read more: https://lnkd.in/ep9zzWJ9

?? Introducing Fireworks f1 ?? A compound AI model specialized in complex reasoning. f1 is the first reasoning system over open models to beat GPT-4o and Claude 3.5 Sonnet across hard coding, chat and math benchmarks. At Fireworks AI, we believe the future of AI is shifting to compound AI systems that combine specialized models and tools to achieve better performance, reliability and control. However building compound AI systems is difficult and time-consuming, so we set out to fix that. Today, we’re releasing a first step in that direction. f1 is a compound AI model specialized in complex reasoning, that interweaves multiple open models at the inference layer. f1 enables developers to access the power of compound AI with the simplicity of prompting. Using prompt as the universal declarative programming language for Gen AI application building, developers can describe what they want to achieve without needing to specify exactly how to accomplish it. ? two variants now available in preview: f1 and f1-mini ? access the preview on Fireworks AI Playground for free ? get on to waitlist for free early access to the f1 API We invite you to help us improve these models and shape the future of compound AI. Read more: https://lnkd.in/ep9zzWJ9

?? ProoferX: A Game-Changer for Reliable Technical Documentation ProoferX tackles a major pain point for developers: outdated code examples in technical documentation. Leveraging Fireworks AI's Llama model endpoints and Firefunction for structured outputs, the project automates the validation of code snippets, ensuring they work seamlessly across different versions, environments, and dependencies. Why ProoferX stood out: ? End-to-End Code Extraction: Analyzes URLs with Firecrawl to pull complete, executable code snippets directly from docs. ? Goal-Oriented Validation: Uses Llama 3.2 models to extract intent and define success criteria for each code snippet. ? Seamless Execution Pipeline: Validates code in sandbox environments with structured data handling, reducing manual checks. Nehil Jain + Selvam Palanimalai built a polished front-end and demo that caught real-world coding errors and broken snippets in various docs and various languages. Their deep understanding of Fireworks features and models (bolstered by Nehil's participation in three Fireworks hackathons) was evident throughout the project. ?? Check out the full story here: https://lnkd.in/eru8pv3M #fireworks #genai #usershowcase #gallery #community #hackathon #e2b #llms

We're #hiring a new Backend Software Engineer in Redwood City, California. Apply today or share this post with your network.