# Vitna: AI intelligence and infrastructure > Understand the AI market. Get useful web data. Run models on your own infrastructure. > Compare models, providers and prices. Search and read the web. Access cloud models or run open models on your own hardware. > Source: https://vitna.ai/ > Mode: Live record read (Tape confirmed 22 Nov 14:08 UTC) ## Verified record summary - Price rows on the tape: 1,572 - Models priced on the tape: 370 - Providers priced: 33 - Dated market changes: 216 (14 Mar 2023 to 19 Nov 2026) - Latest signed day: 42,528 prices on 21 Nov 2026 (root bb8b7adf3165, sha256-merkle-rfc6962, Ed25519) - Recent attestation window: 30 of the last 30 days signed - Register inventory: 713 skus on the register; 451 listings in OpenRouter's catalog at the 17 Nov 16:29 UTC read; 34 providers on the register ## Flagship model prices (USD per 1M tokens) | Model | Maker | Input | Output | Cached input | | :--- | :--- | :--- | :--- | :--- | | Claude Opus 5.5 | Anthropic | $4.00 | $20.00 | $0.20 | | GPT-6 Astra | OpenAI | $10.00 | $50.00 | $1.00 | | GPT-6.1 Sol | OpenAI | $2.00 | $10.00 | $0.10 | | Claude Sonnet 5.5 | Anthropic | $2.00 | $10.00 | $0.10 | | Gemini 4 Argon | Google | $2.00 | $10.00 | $0.10 | | Grok 4.7 | SpaceXAI | $2.00 | $6.00 | $0.50 | ## Latest market changes - **19 Nov**: Novita AI Kimi K3 cache write price, 1 hour retention falls to $0.30 - **19 Nov**: Novita AI Kimi K3 cache write price, 5 minute retention rises to $6.00 - **19 Nov**: Novita AI Kimi K3 cached input price rises to $3.00 - **19 Nov**: Novita AI Qwen3.8-Max cache write price, 5 minute retention falls to $0.25 - **19 Nov**: Novita AI Qwen3.8-Max cached input price rises to $2.50 - **19 Nov**: Novita AI Qwen3.8-Flash cache write price, 5 minute retention falls to $0.016 - **19 Nov**: Novita AI Qwen3.8-Flash cached input price rises to $0.20 - **16 Nov**: Together AI Qwen Image 2.0 per image falls to $0.035 ## Nine-layer stack - **09 Apps**: Chat, research and agents, as software a person opens. - **08 Surfaces**: Where people and software meet the stack. - **07 Accounts and keys**: Sign in, open an organisation, add people, mint keys, pay. - **06 Decisions**: Decide when a production stack should change, show the arithmetic, and let a person rule. - **05 Evidence**: Every step from a published price to a finished run, signed, and pointing at the one before. - **04 Spend and permission** (In development): An agent spends and acts only inside authority somebody delegated to it. - **03 Agents and tools**: Multi-step work with tools, in a sandbox, with approvals. - **02 The gateway**: One OpenAI-compatible endpoint on your side, serving the decision per request. - **01 Models and inference**: Reach models on your own keys, in the cloud or on your own hardware. ## Quickstart commands ```bash # Keyless MCP server (Claude Code and Streamable HTTP clients) claude mcp add --transport http vitna https://vitna.ai/api/v1/mcp # Public price tape (no key required) curl https://vitna.ai/api/v1/markets/model/tape # Self-hosted local gateway (Apache-2.0, zero inference egress to vitna.ai) npm install -g vitna-runtime vitna-relay # Self-hosted container image docker pull ghcr.io/convexityos/vitna-runtime ``` ## Plans Structural neutrality holds across every tier: Vitna charges for hosted web-data capacity and decision control-plane software, never a spread on LLM tokens or per-seat fees. | Plan | Price | Active workloads | Web API pages/day | Search/day | Unblocker/day | | :--- | :--- | :--- | :--- | :--- | :--- | | Free | $0 (no card, no expiry) | 1 | 300 | 250 | 100 | | Builder | $29/month | 3 | 5,000 | 2,500 | 1,000 | | Pro | $79/month | 5 | 25,000 | 15,000 | 3,000 | | Team | $299/month | 25 | 100,000 | 50,000 | 12,000 | | Enterprise | Custom | Custom | Custom | Custom | Custom | ## Machine discovery - Corpus index: https://vitna.ai/llms.txt - Full corpus: https://vitna.ai/llms-full.txt - OpenAPI 3.1: https://vitna.ai/openapi.json - Unified MCP manifest: https://vitna.ai/.well-known/mcp.json - A2A agent card: https://vitna.ai/.well-known/agent-card.json - Keyless market MCP endpoint: https://vitna.ai/api/v1/mcp ## Site directory ### Intelligence - [Overview](https://vitna.ai/intelligence) - [Models](https://vitna.ai/models) - [Compare](https://vitna.ai/compare) - [Benchmarks](https://vitna.ai/benchmarks) - [Rankings](https://vitna.ai/rankings) - [Prices](https://vitna.ai/markets) - [Market changes](https://vitna.ai/log) - [Providers](https://vitna.ai/providers) - [Today](https://vitna.ai/today) - [Decision Lab](https://vitna.ai/lab) ### Web data - [Search](https://vitna.ai/search) - [Web APIs](https://vitna.ai/apis) - [Unblocker](https://vitna.ai/unblocker) - [Compare web data](https://vitna.ai/infrastructure#web-data) ### Model infrastructure - [Relay](https://vitna.ai/relay) - [Runtime](https://vitna.ai/runtime) - [Anchor](https://vitna.ai/anchor) - [Compare model infra](https://vitna.ai/infrastructure#models) ### Developers - [Start here](https://vitna.ai/developers) - [API workbench](https://vitna.ai/apis#try) - [Documentation](https://vitna.ai/docs) - [Public API](https://vitna.ai/docs/api/v1/overview) - [Web API reference](https://vitna.ai/docs/web/overview) - [Search reference](https://vitna.ai/docs/search/overview) - [Runtime and Relay](https://vitna.ai/docs/runtime/overview) - [MCP servers](https://vitna.ai/mcp) - [Invoice reconcile](https://vitna.ai/reconcile) - [Keys and accounts](https://vitna.ai/developers#keys) - [OpenAPI specs](https://vitna.ai/openapi.json) - [llms.txt](https://vitna.ai/llms.txt) ### Company and trust - [Methodology](https://vitna.ai/method) - [Coverage](https://vitna.ai/coverage) - [Verification](https://vitna.ai/docs/verify-a-figure) - [Citing a figure](https://vitna.ai/docs/citing-a-figure) - [The collector](https://vitna.ai/witness) - [Case study](https://vitna.ai/case) - [Posts](https://vitna.ai/posts) - [Router Challenge](https://vitna.ai/challenge) - [Workspace concept](https://vitna.ai/platform) - [Brand](https://vitna.ai/brand)