# Ambient

> Ambient is a Proof of Work AI blockchain and verified inference network. It runs top open models on a decentralized network of GPUs and lets any user, application, or agent verify - via Proof of Logits - that a specific model ran on a specific prompt and produced a specific output.

Ambient is verified AI infrastructure, not a generic model marketplace. Miners perform useful work (inference and model improvement) instead of arbitrary hashing, and the network verifies that work so it can be trusted. Public content on this site is intentionally available for search, AI input, and AI training.

## Core Context

- [Ambient overview](https://ambient.xyz/ambient-overview.md): Canonical description of Ambient, useful Proof of Work, and Proof of Logits.
- [Verified inference](https://ambient.xyz/verified-inference.md): Why verified inference matters for LLMs, agents, enterprises, and consumers.
- [Agentic economy](https://ambient.xyz/agentic-economy.md): Why autonomous agents need provable AI execution.
- [Enterprise AI](https://ambient.xyz/enterprise-ai.md): The enterprise value of API-compatible verified inference.
- [FAQ](https://ambient.xyz/faq.md): Concise answers to common questions about Ambient.
- [Full LLM context](https://ambient.xyz/llms-full.txt): Single-file context bundle for models and agents.

## Products

- [Models & API pricing](https://ambient.xyz/models): 3+ open models including Kimi K2.7 Code and GLM 5.2 with per-token pricing.
- [Pricing & plans](https://ambient.xyz/pricing): Subscription plans and pay-as-you-go verified inference.
- [Coding plan](https://ambient.xyz/coding-plan): Verified open models for OpenAI/Anthropic-compatible coding tools.
- [Enterprise](https://ambient.xyz/enterprise): Verifiable, private, auditable inference for enterprises.
- [Mining](https://ambient.xyz/mining): Contribute GPU compute and earn rewards for verified work.

## Docs

Every documentation page is also published as plain Markdown — append ".md" to any docs URL.

- [Get started](https://ambient.xyz/docs/get-started.md): Point any OpenAI- or Anthropic-compatible client at Ambient's API and make your first request against the verified inference network.
- [Quickstart](https://ambient.xyz/docs/get-started/quickstart.md): Create an API key, make your first request against the Ambient network, and understand what comes back, all in under five minutes.
- [Using the web app](https://ambient.xyz/docs/get-started/web-app.md): Sign in at app.ambient.xyz to chat with the network model, run Research mode, generate API keys, monitor usage, manage teams, claim Drops, and handle billing.
- [Build on Ambient](https://ambient.xyz/docs/build.md): The Ambient API surface for developers. OpenAI-compatible chat completions and Anthropic-compatible messages at one base URL, with bearer-key auth.
- [Coding agents](https://ambient.xyz/docs/build/coding-agents.md): Bring Ambient's verified open models into the coding agent you already use, whether Claude Code, Codex, OpenCode, or OpenClaw. Keep your agent as the orchestrator and route the token-heavy work to the network, or run the whole session on Ambient.
- [Claude Code](https://ambient.xyz/docs/build/claude-code.md): Use Ambient from Claude Code. Install the ambient-code plugin so Claude routes the token-heavy building to Ambient's open models while it orchestrates, or point Claude Code's own model at Ambient to run the whole session on the network.
- [Codex](https://ambient.xyz/docs/build/codex.md): Bring Ambient into Codex with the ambient-codex plugin. Route the bulk of your coding to Ambient's open models with Delegate and Ambient-session (takeover) modes, while Codex stays the control plane for review, tests, and the final commit.
- [OpenCode](https://ambient.xyz/docs/build/opencode.md): Configure OpenCode to use Ambient as its inference provider by selecting Ambient from the built-in provider menu and pasting your API key.
- [OpenClaw](https://ambient.xyz/docs/build/openclaw.md): Configure OpenClaw to use Ambient as its inference provider by installing the Ambient plugin from ClawHub and onboarding with your API key.
- [OpenAI SDK](https://ambient.xyz/docs/build/openai-sdk.md): Use the official OpenAI SDK with Ambient. Point base_url at api.ambient.xyz/v1, then chat and stream exactly as you would against OpenAI.
- [Anthropic SDK](https://ambient.xyz/docs/build/anthropic-sdk.md): Use the official Anthropic SDK with Ambient. Point base_url at api.ambient.xyz and use the Messages API, including thinking blocks and streaming.
- [Models and readiness](https://ambient.xyz/docs/build/models.md): The Ambient model catalog. How to read the is_ready flag, what the ambient/large alias is, and why pricing always comes from the API.
- [Streaming](https://ambient.xyz/docs/build/streaming.md): Stream completions over SSE. Standard OpenAI-style chunks, the dual reasoning/reasoning_content delta keys, and how to detect stalls and truncation.
- [Errors and retries](https://ambient.xyz/docs/build/errors.md): The Ambient API error taxonomy. What each status code actually means, why 429 "No workers available" is not a rate limit, and safe retry rules.
- [Verified inference](https://ambient.xyz/docs/build/verified-inference.md): The verification surface of the Ambient API, covering request flags and the verified and merkle_root response fields.
- [API reference](https://ambient.xyz/docs/build/api.md): The full Ambient HTTP API surface, covering base URL, auth, compatibility dialects, and a generated table of every endpoint in the live OpenAPI spec.
- [Chat Completions](https://ambient.xyz/docs/build/api/chat-completions.md): POST /v1/chat/completions field by field. The OpenAI-compatible request schema, the Ambient extension flags, and the response shape.
- [Models endpoint](https://ambient.xyz/docs/build/api/models-endpoint.md): GET /v1/models reference. The unauthenticated catalog endpoint, per-model pricing and capability fields, and what is_ready actually tells you.
- [Pay per request with x402](https://ambient.xyz/docs/build/x402.md): Pay-per-request inference through JumpGate and the x402 payment protocol. Quote an exact request, pay in USDC on Solana or Base, and stream the response.
- [Headless x402 subscriptions](https://ambient.xyz/docs/build/x402-subscriptions.md): Register, obtain and validate a live offer, authorize and pay, complete provisioning, use the account, and renew an API-only Ambient subscription headlessly over x402.
- [Restart-safe x402 subscriptions in TypeScript](https://ambient.xyz/docs/build/x402-subscriptions-example.md): A durable-state TypeScript reference implementation of the Ambient x402 subscription flow that survives a process restart at every stage, from unsigned challenge to completed provisioning.
- [The auction program](https://ambient.xyz/docs/build/on-chain.md): Ambient's on-chain auction program. JobRequest lifecycle, bundle and auction account states, the second-lowest reverse auction, and how to request a verified inference job via CPI.
- [Tool Oracle](https://ambient.xyz/docs/build/tool-oracle.md): On-chain LLM inference calls. The Tool Oracle lets you make LLM inference calls with on-chain inputs and configuration, regex-filter outputs, and write results back on-chain where your programs can consume them.
- [Solana quickstart](https://ambient.xyz/docs/build/solana-quickstart.md): Connecting Solana tooling to Ambient. The Solana-compatible RPC endpoint, SVM compatibility, and where the JumpGate relayer fits.
- [Use Ambient](https://ambient.xyz/docs/use.md): What you can use today. The web app at app.ambient.xyz for chat, research, keys, and billing, plus Ambient Desktop.
- [Ambient web app](https://ambient.xyz/docs/use/app.md): The three surfaces of the Ambient web app. Chat for conversation, Research for citation-rich analysis, and API Keys for integrating Ambient into your own code.
- [Ambient Desktop](https://ambient.xyz/docs/use/desktop.md): The open agentic workstation. A local-first agentic workstation, model-orchestration engine, and miner that runs capable agents on your real machine instead of a quarantined VM.
- [Product north star](https://ambient.xyz/docs/use/desktop/north-star.md): The candid product thesis. Ambient Desktop becomes the workstation for agentic work on the Ambient Network, with coding as the first high-value application and four product surfaces on the path there.
- [Install Ambient Desktop](https://ambient.xyz/docs/use/desktop/installation.md): Install Ambient Desktop from packaged artifacts on macOS, Windows, and Linux, run the per-capability health checks after first launch, or build it from source.
- [Desktop quickstart](https://ambient.xyz/docs/use/desktop/quickstart.md): Start in the pinned Welcome Folder, finish Core and Plugin Setup, then run one small calculator task and inspect the proof it leaves behind.
- [Project Board](https://ambient.xyz/docs/use/desktop/project-board.md): Turn large or ambiguous goals into a Kanban of source-backed cards with evidence, dependencies, and a review loop instead of a transcript you scroll.
- [Durable goals](https://ambient.xyz/docs/use/desktop/durable-goals.md): Long-running goals persist their plan, evidence, and continuation state to disk, so work survives restarts and context limits instead of dissolving into a chat transcript.
- [Symphony](https://ambient.xyz/docs/use/desktop/symphony.md): Multi-agent orchestration: a parent session owns the objective, budget, and approvals, and delegates scoped work to typed child agents joined back through artifacts.
- [Workflow Recorder](https://ambient.xyz/docs/use/desktop/workflow-recorder.md): Record a successful multi-step task once, review and generalize it with Ambient, and re-run it as a parameterized, callable workflow.
- [Provider routing](https://ambient.xyz/docs/use/desktop/provider-routing.md): Ambient orders search, fetch, browser, vision, and local-model providers by your preference and task fit, with policy checks and visible fallback evidence when one is skipped.
- [Remote control](https://ambient.xyz/docs/use/desktop/remote-control.md): Drive a running Ambient Desktop from Telegram, owner-authenticated and preview-then-approve, with anti-replay protection. Check status, change settings, and run work away from the window.
- [Security model](https://ambient.xyz/docs/use/desktop/security.md): Ambient Desktop sandboxes the capability, not the agent, with ToolHive containers, descriptor review, permission profiles, secret redaction, and audit artifacts, while the agent keeps working on your real desktop.
- [Curated defaults](https://ambient.xyz/docs/use/desktop/curated-defaults.md): Ambient Desktop ships a small set of reputable, reviewed, pinned, and sandboxed capabilities as its default catalog, so the common tasks work without an agent installing unvetted packages.
- [Browser and scraping safety](https://ambient.xyz/docs/use/desktop/browser-scraping.md): Ambient Desktop treats every fetched page as untrusted input. URL policy, private-address protection, isolated runtimes, bounded outputs, and visible approval bound what a web capability can do.
- [Privacy and data retention](https://ambient.xyz/docs/use/desktop/privacy.md): Ambient's zero-data-retention guarantee. What is enforced today on testnet, and what onion routing and end-to-end encryption add on mainnet.
- [Network client](https://ambient.xyz/docs/use/desktop/network-client.md): Ambient Desktop is growing into a first-party client for the Ambient network, with wallet, balances, programs, and transaction evidence. This surface is in active development.
- [Local models and routing](https://ambient.xyz/docs/use/desktop/local-models.md): How Ambient Desktop routes work across local llama.cpp runtimes, network models, and cloud providers, keeping private, cheap, or latency-sensitive work on your machine.
- [Plugins and marketplace](https://ambient.xyz/docs/use/desktop/plugins.md): How Ambient Desktop extends safely. Four install classes with visible permissions, the Plugin Manager, a Capability Builder for custom tools, and a curated marketplace direction.
- [Local visual analysis](https://ambient.xyz/docs/use/desktop/visual-analysis.md): Ambient Desktop routes screenshots and images to a MiniCPM-V vision model on your own machine, so text models without native vision can still see, no cloud round trip.
- [Agent memory](https://ambient.xyz/docs/use/desktop/agent-memory.md): An experimental, opt-in memory system that recalls and captures context across turns, stored workspace-local in SQLite, off by default, inspectable and deletable.
- [Desktop API endpoints](https://ambient.xyz/docs/use/desktop/api-endpoints.md): The OpenAI-compatible inference API that Ambient Desktop and any other agent point at. Base URL, auth header, model IDs, and the endpoints you call.
- [Ambient Desktop vs Codex CLI](https://ambient.xyz/docs/use/desktop/comparisons/codex-cli.md): Both ship a coding agent, so the comparison turns on form factor. A terminal-native loop versus a desktop workspace with persistent plans, sandboxed capabilities, and reviewable evidence.
- [Ambient Desktop vs Hermes](https://ambient.xyz/docs/use/desktop/comparisons/hermes.md): Both extend through MCP, so the comparison turns on where the agent lives. Always-on autonomy reachable from messaging platforms versus a local desktop workstation you drive and inspect.
- [Ambient Desktop vs open marketplaces](https://ambient.xyz/docs/use/desktop/comparisons/openclaw.md): The comparison turns on default trust posture. An open feed of community plugins versus a curated, sandboxed default set that you extend deliberately.
- [Learn](https://ambient.xyz/docs/learn.md): The concepts behind Ambient: what the network is, how Proof of Logits verifies inference, how requests flow, and where economics stand.
- [What is Ambient?](https://ambient.xyz/docs/learn/what-is-ambient.md): Ambient is an SVM-compatible proof-of-work L1 that serves verified AI inference from one continuously improving, open-weights foundation model.
- [Proof of Logits](https://ambient.xyz/docs/learn/proof-of-logits.md): How Ambient verifies that a specific model produced a specific output: logit fingerprints, progress markers, and cheap spot-check validation.
- [Network architecture](https://ambient.xyz/docs/learn/architecture.md): The roles that make up the Ambient network (users, the gateway layer, miners, validators) and how a request flows from auction to settlement.
- [Testnet](https://ambient.xyz/docs/learn/testnet.md): The live Ambient testnet (external services, the on-chain Ambient Main Program, and the off-chain miner/validator components) plus how developers, miners, and validators take part.
- [Tokenomics](https://ambient.xyz/docs/learn/tokenomics.md): How Ambient rewards validated useful work. Work credit, work-weighted leader election, and auction pricing.
- [Run a node](https://ambient.xyz/docs/mine.md): How mining works on Ambient today: live, curated onboarding waves for datacenter-class hardware, with a consumer Mini Miner on the roadmap.
- [Mini Miner](https://ambient.xyz/docs/mine/mini-miner.md): The planned, opt-in path for a machine already running local models to serve the Ambient network and earn rewards. Roadmap, nothing to install or buy today.
- [Enterprise](https://ambient.xyz/docs/enterprise.md): Ambient for organizations: verified inference for auditability, data retention and privacy, and how to start a deployment conversation.
- [Maturity labels](https://ambient.xyz/docs/resources/maturity-labels.md): What Available, Developer Preview, In Development, and Roadmap each promise, and how they map to the maturity value every page in these docs carries.
- [Changelog](https://ambient.xyz/docs/resources/changelog.md): Dated changes to the Ambient network and this documentation.

## Optional

- [Ambient app](https://app.ambient.xyz): Web chat, desktop, and API keys.
- [Ambient docs](https://ambient.xyz/docs): Developer documentation.
- [Ambient on X](https://x.com/ambient_xyz): Public updates.