# Narev Docs

## Docs

- [AI adoption rate for large firms continues to trend down](https://narev.ai/docs/blog/ai-adoption-rate.md): U.S. Census Bureau data shows AI adoption among large firms has continued to decline after peaking in July 2025, while smallest firms keep growing
- [GPT-3.5 MMLU Score: why it beats GPT-5 at 3% of the Cost](https://narev.ai/docs/blog/gpt35-beats-gpt5.md): Despite its low MMLU score, GPT-3.5 outperformed top 2025 models like GPT-5 and Claude Opus on a real task, costing $823 vs $30,390 per 1M requests.
- [Narev Blog](https://narev.ai/docs/blog/index.md): Latest news, research, and updates on AI economics and FinOps.
- [GPT-4o vs Claude Opus - MMLU Scores vs. Actual API Costs](https://narev.ai/docs/blog/mmlu-doesnt-matter.md): MMLU scores for GPT-4o vary by 13 points, while top models differ by 1%. The real difference? Massive cost disparities for the same performance.
- [Why Narev AI Exists - Two futures for AI economics](https://narev.ai/docs/blog/why-we-launch.md): There's a world where AI gets infinitely cheaper, and another where users need to choose the best tool for every job. Narev is building for the latter.
- [DeepSeek Usage-Based Billing with Vercel AI SDK](https://narev.ai/docs/guides/deepseek-usage-based-billing.md): Learn how to track usage and calculate costs for DeepSeek models using the @ai-billing/deepseek middleware.
- [DeepSeek V4 Models Pricing](https://narev.ai/docs/guides/deepseek-v4-pricing.md): Learn how to look up and calculate pricing for DeepSeek V4 models using the Narev API.
- [FinOps for AI Framework](https://narev.ai/docs/guides/finops-for-ai/index.md): A practical 3-step framework to measure, track, and optimize your LLM spending without compromising quality.
- [Step 1: know your objective and who's in charge](https://narev.ai/docs/guides/finops-for-ai/step-1.md): Establish what success looks like before optimizing. Get your team aligned on goals and identify who makes the final call.
- [Step 2: know what you're spending on](https://narev.ai/docs/guides/finops-for-ai/step-2.md): Track LLM costs at the source. Break down aggregate spending into actionable insights by app, feature, and team.
- [Step 3: optimize](https://narev.ai/docs/guides/finops-for-ai/step-3.md): Test systematically. Measure ruthlessly. Deploy winners confidently. Here's how to cut costs without sacrificing quality.
- [How to choose an LLM](https://narev.ai/docs/guides/how-to-choose-llm-model.md): Guide to selecting the best LLM for your product
- [How to choose an LLM (with labelled data)](https://narev.ai/docs/guides/how-to-choose-llm-model-with-labels.md): Guide to selecting the best LLM for your product (with labelled data)
- [Guides and Best Practices](https://narev.ai/docs/guides/index.md): Learn how to optimize your AI stack for cost, performance, and reliability.
- [Reduce LLM spend by switching models](https://narev.ai/docs/guides/reduce-cost-by-model-switch.md): How switching from GPT-4 to gpt-oss-20b cut costs by 99% while maintaining 100% accuracy
- [Reduce LLM spend by prompt engineering](https://narev.ai/docs/guides/reduce-cost-by-prompt-engineering.md): How a shorter, simpler prompt cut costs by 24% with the same accuracy and better consistency
- [Welcome to Narev](https://narev.ai/docs/index.md): The complete AI FinOps ecosystem for tracking, benchmarking, and billing.
- [Amazon Web Services](https://narev.ai/docs/narev-oss/connect-providers/aws.md): Connect AWS billing exports to Narev Self-Hosted using S3 and Identity and Access Management (IAM) credentials.
- [Azure](https://narev.ai/docs/narev-oss/connect-providers/azure.md): Connect Azure Cost Management exports from Blob Storage to Narev Self-Hosted.
- [Google Cloud Platform](https://narev.ai/docs/narev-oss/connect-providers/gcp.md): Connect GCP BigQuery billing exports with a FOCUS view to Narev Self-Hosted.
- [Overview](https://narev.ai/docs/narev-oss/connect-providers/index.md): Connect supported cloud and AI providers to start importing billing data.
- [Open AI](https://narev.ai/docs/narev-oss/connect-providers/openai.md): Connect OpenAI usage and billing data with an administrator API key.
- [FOCUS Specification](https://narev.ai/docs/narev-oss/focus-specification.md): Understand the FOCUS format fields and categories used by Narev Self-Hosted.
- [Getting Started](https://narev.ai/docs/narev-oss/getting-started/deployment.md): Deploy Narev Self-Hosted with Docker and configure production settings.
- [Sync Providers](https://narev.ai/docs/narev-oss/getting-started/sync-providers.md): Manually sync connected providers and troubleshoot billing data imports.
- [Welcome to narevai/narev](https://narev.ai/docs/narev-oss/index.md): Set up Narev Self-Hosted to unify cloud and AI billing data in FOCUS format.
- [Chat completions](https://narev.ai/docs/platform/api-reference/endpoint/applications/chat-completions.md): Send chat completion requests to an A/B test using an OpenAI-compatible endpoint.
- [Submit custom metrics](https://narev.ai/docs/platform/api-reference/endpoint/applications/custom-metrics.md): Submit a custom quality metric value for a response generated through the Applications API.
- [Calculate cost for a model call](https://narev.ai/docs/platform/api-reference/endpoint/pricing/calculate-cost-for-a-model-call.md): Given a model ID, gateway, provider, and token usage, returns an itemized cost breakdown in USD.
- [List pricing for models](https://narev.ai/docs/platform/api-reference/endpoint/pricing/list-model-pricing.md): Returns a map of model IDs to provider pricing rows. All query parameters are optional. Omit them to fetch the full catalog.
- [Chat completions](https://narev.ai/docs/platform/api-reference/endpoint/router/chat-completions.md): Route chat completion requests to A/B tests using rules and filters configured in the Narev Cloud dashboard.
- [Introduction](https://narev.ai/docs/platform/api-reference/introduction.md): HTTP APIs for AI model pricing, A/B test applications, and intelligent request routing on Narev Cloud.
- [Analyzing Results](https://narev.ai/docs/platform/benchmark/analyzing-results.md): Understand your A/B test results and compare variant performance.
- [Create benchmark](https://narev.ai/docs/platform/benchmark/create.md): Set up a benchmark to organize your prompts and compare variants.
- [Clawhub](https://narev.ai/docs/platform/benchmark/data-source/clawhub.md): Set up a benchmark to organize your prompts and compare variants.
- [File Upload](https://narev.ai/docs/platform/benchmark/data-source/file-upload.md): Set up a benchmark to organize your prompts and compare variants.
- [Live Test](https://narev.ai/docs/platform/benchmark/data-source/live-test.md): Set up a benchmark to organize your prompts and compare variants.
- [Manual entry](https://narev.ai/docs/platform/benchmark/data-source/manual-entry.md): Set up a benchmark to organize your prompts and compare variants.
- [Tracing platform](https://narev.ai/docs/platform/benchmark/data-source/tracing-platform.md): Set up a benchmark to organize your prompts and compare variants.
- [Helicone](https://narev.ai/docs/platform/benchmark/integration/helicone.md): Import production traces from Helicone into Narev.
- [Helicone](https://narev.ai/docs/platform/benchmark/integration/helicone-gateway.md): Test model configurations in Narev using production data from Helicone Gateway.
- [Integration overview](https://narev.ai/docs/platform/benchmark/integration/index.md): Narev complements your observability, gateways, and FinOps tools
- [Langfuse](https://narev.ai/docs/platform/benchmark/integration/langfuse.md): Import production traces from Langfuse into Narev.
- [LangSmith](https://narev.ai/docs/platform/benchmark/integration/langsmith.md): Import production traces from LangSmith into Narev.
- [LiteLLM](https://narev.ai/docs/platform/benchmark/integration/litellm.md): Test LiteLLM gateway configurations in Narev using production data.
- [OpenAI](https://narev.ai/docs/platform/benchmark/integration/openai.md): Test OpenAI models in Narev using production data.
- [OpenRouter](https://narev.ai/docs/platform/benchmark/integration/openrouter.md): Test OpenRouter models in Narev using production data.
- [Portkey](https://narev.ai/docs/platform/benchmark/integration/portkey.md): Test Portkey gateway configurations in Narev using production data.
- [Weights & Biases Weave](https://narev.ai/docs/platform/benchmark/integration/wandb.md): Import production traces from W&B Weave into Narev.
- [Introduction](https://narev.ai/docs/platform/benchmark/introduction.md): Learn how to compare variants and optimize your AI applications with A/B testing in Narev.
- [Benchmarking with API](https://narev.ai/docs/platform/benchmark/using-api.md): Use the Applications API to run A/B tests programmatically.
- [Adding Variants](https://narev.ai/docs/platform/benchmark/variant/adding-variants.md): Each benchmark requires at least one variant
- [Creating Variants](https://narev.ai/docs/platform/benchmark/variant/creating-variants.md): Configure model variants with different configurations to compare in your benchmarks
- [Benchmarks](https://narev.ai/docs/platform/concepts/benchmarks.md): Datasets that hold your prompts and let you compare variants side by side.
- [Quality evaluations](https://narev.ai/docs/platform/concepts/quality-evaluations.md): Define how to evaluate whether a variant performed correctly.
- [Routers](https://narev.ai/docs/platform/concepts/routers.md): Send queries to the right variants.
- [Variants](https://narev.ai/docs/platform/concepts/variants.md): Your model configuration: model choice, system prompt, and parameters.
- [Why benchmark?](https://narev.ai/docs/platform/concepts/why-benchmark.md): Learn why academic benchmarks aren't enough and why your own product data should drive model choices.
- [Why router?](https://narev.ai/docs/platform/concepts/why-router.md): Understand how routing differs from gateways and how sending each prompt to the right model cuts cost without sacrificing quality.
- [Quickstart: benchmarks](https://narev.ai/docs/platform/quickstart/benchmark.md): Add a model variant to a benchmark and compare quality and cost.
- [Quickstart: routers](https://narev.ai/docs/platform/quickstart/router.md): Create a sequential fallback router and use it in your app.
- [Filter Routers](https://narev.ai/docs/platform/routing/filter-routers.md): Learn how to configure routers with routing rules, filter conditions, and fallback destinations.
- [Introduction](https://narev.ai/docs/platform/routing/introduction.md): Set up routers to intelligently route requests to different A/B tests based on dynamic conditions.
- [Sequential Routers](https://narev.ai/docs/platform/routing/sequential-routers.md): Learn how to configure routers with routing rules, filter conditions, and fallback destinations.
- [Routing with the API](https://narev.ai/docs/platform/routing/using-api.md): Use the Router API to dynamically route requests to different A/B tests.
- [Narev SDKs overview](https://narev.ai/docs/sdk/index.md): Libraries that make billing for AI easy—drop them into your app and start metering in minutes

## OpenAPI Specs

- [openapi](https://narev.ai/docs/platform/api-reference/openapi.json)

## Optional

- [Guides](https://narev.ai/guides)
- [Blog](https://narev.ai/blog)