llms.txt
Compare with other sites' llms.txt files →# Fireworks AI
> Fireworks is a training and inference platform for open AI models. Teams use it to serve frontier open models (or their own post-trained versions) on a disaggregated inference engine, and to post-train models with SFT, DPO, and reinforcement learning, then deploy checkpoints in one click.
Key facts:
- Inference options: Serverless (per-token pricing), On-Demand deployments, and Reserved Capacity.
- Training options: Training API, Managed Training, and Fireworks Lab (SFT, DPO, RL).
- Nexus lets teams run open models in coding agents such as Claude Code, Codex, and OpenCode, with FireRouter routing, per-user budgets, and spend tracking by model.
- Developer documentation lives at https://docs.fireworks.ai, with its own index at https://docs.fireworks.ai/llms.txt.
## Product
- [Inference](https://fireworks.ai/product/inference): Serve open models or your post-trained versions on a disaggregated engine with Serverless, On-Demand, or Reserved Capacity
- [Training](https://fireworks.ai/product/training): Post-train open models with SFT, DPO, or RL, then deploy checkpoints in one click
- [RL Rollouts](https://fireworks.ai/product/rl-rollouts): Standalone service for large-scale reinforcement learning rollouts
- [Nexus](https://fireworks.ai/product/nexus): Control AI coding spend with open models in Claude Code, Codex, and OpenCode
- [Pricing](https://fireworks.ai/pricing): Plans and pricing
- [Fireworks Specialized Intelligence Index](https://fireworks.ai/specialized-intelligence-index/): How open, closed, and specialized models perform on the evaluations practitioners use
## Solutions
- [Fireworks for enterprise](https://fireworks.ai/solutions/for-enterprise)
- [Fireworks for startups](https://fireworks.ai/solutions/for-startups)
- [Agentic Systems](https://fireworks.ai/solutions/agentic-systems): Real-time, reliable execution of multi-agent workflows across systems and APIs
- [Code Assistance](https://fireworks.ai/solutions/code-assistance): Context-aware code generation, inline fixes, and real-time autocomplete
- [Enterprise RAG](https://fireworks.ai/solutions/enterprise-rag): Retrieval-augmented generation over enterprise documents, code, and data
- [Voice Agent Platform](https://fireworks.ai/solutions/voice-agent-platform): Build fast, configurable voice agents
- [Reinforcement Fine-Tuning](https://fireworks.ai/solutions/reinforcement-fine-tuning)
- [Multimodal](https://fireworks.ai/solutions/multimodal): Extract, classify, and summarize documents, images, and audio
- [Search](https://fireworks.ai/solutions/search): Context-aware search and multi-source summarization
- [Conversational AI](https://fireworks.ai/solutions/conversational-ai): Domain-specific conversational AI for enterprise workflows
## Models
- [Models](https://fireworks.ai/models): Serve frontier open models, or your own post-trained versions, on an engine optimized at every layer
- [Kimi on Fireworks](https://fireworks.ai/models/kimi): Moonshot AI's Kimi models via serverless API or dedicated GPUs
- [DeepSeek on Fireworks](https://fireworks.ai/models/deepseek): DeepSeek models for inference and fine-tuning
- [Which model should I use?](https://docs.fireworks.ai/guides/recommended-models.md): Pick the best open-weight model for chat, coding, vision, embeddings, and agents
## Docs
- [Fireworks AI Docs index](https://docs.fireworks.ai/llms.txt): Full documentation index
- [Build with Fireworks AI](https://docs.fireworks.ai/getting-started/introduction.md)
- [Serverless Quickstart](https://docs.fireworks.ai/getting-started/quickstart.md)
- [Serverless Pricing](https://docs.fireworks.ai/serverless/pricing.md): Per-token serverless pricing
## Customers and resources
- [Customers](https://fireworks.ai/customers): Case studies from companies building on Fireworks
- [Blog](https://fireworks.ai/blog): Announcements, model releases, perspectives, and tutorials
- [Research](https://fireworks.ai/research): Research on training, inference, and model performance
- [Newsroom](https://fireworks.ai/news): Latest news from Fireworks
- [AI Tokenomics Whitepaper](https://fireworks.ai/tokenomics): Why AI economics are set at the model layer, and what to rent vs. own
- [Own your AI whitepaper](https://fireworks.ai/own-your-ai)
## Events
- [Events](https://fireworks.ai/events): Upcoming events and conferences
- [Webinars](https://fireworks.ai/webinar)
- [Community](https://fireworks.ai/community): Meetups, nerd nights, hackathons, and workshops
## Partners
- [Partners](https://fireworks.ai/partners)
- [NVIDIA](https://fireworks.ai/partners/nvidia)
- [AMD](https://fireworks.ai/partners/amd)
- [AWS](https://fireworks.ai/partners/aws)
- [Microsoft Foundry](https://fireworks.ai/partners/microsoft)
## Company
- [About](https://fireworks.ai/about)
- [Careers](https://fireworks.ai/careers)
- [Contact](https://fireworks.ai/contact): Talk to sales
- [Security bulletins](https://fireworks.ai/security/bulletins)
- [Trust Center](https://trust.fireworks.ai/)
## Optional
- [Privacy Policy](https://fireworks.ai/privacy-policy)
- [Full documentation text](https://docs.fireworks.ai/llms-full.txt)