OpenRouter · 官方公告

Admins can now set organization-wide limits, give people different budgets and model access within them, and change each policy on its own. Existing guardrails keep working.

Our default routing used to send most traffic to the cheapest provider for a model. The new routing algorithm constantly monitors the performance of every provider endpoint to optimize the balance of cost, speed, and reliability for each new request.

OpenRouter will screen all inference requests for child sexual abuse material (CSAM) as they pass through our service. Screening happens in flight, during request processing. Content flagged as potential CSAM may be retained for human review, with reporting and preservation where required by law.

Private Endpoints let Enterprise customers route their own fine-tuned models and dedicated deployments through OpenRouter, alongside the public catalog. Private endpoints are only visible within your organization, and requests go through the API your team already uses. Available by request on the Enterprise plan.

ElevenLabs is now on OpenRouter with 9 Text to Speech and 2 Speech to Text models. Deliver speech with human-like emotion with Eleven v4, give your agent low-latency replies with v4 Turbo, and turn meetings into notes that show who said what with Scribe v2. Here's which model to pick, and one project built start to finish.

We now benchmark model routers side by side across six suites. Our Router Index weighs quality, speed, and cost so you can decide if you need a router and pick the best one for your needs.

We audited our own OpenRouter account and found over 1,000 active API keys for 85 employees, 168 of them hadn't been used in months. The new Security Center helps you manage and de-risk your API keys with one place to see every key, know which ones are risky and why, clean them up, and lock them to your own network.

Send a whole workload in one POST, collect the results within 24 hours, and typically pay half the per-token price. Across 230k+ batches that completed over our two week beta period, the median finished in 7 minutes.

US In-Region Routing is live, joining the EU. Send requests to us.openrouter.ai and their prompts and completions are decrypted only inside the United States and routed only to shared provider endpoints running there, across models from OpenAI, Anthropic, Google, NVIDIA, Thinking Machines, DeepSeek, Moonshot, and Qwen. Requests served by your own BYOK key go to your deployment's region, which you control.

The shell server tool gives any model on OpenRouter a hosted Linux shell, and the Files API moves files in and out of it. Commands run in an isolated container, the model reads back stdout, stderr, and exit codes, and files the run writes can be downloaded afterward. Available today in beta on the Responses and Messages APIs.

We ran 39 image models through 15 deliberately hard prompts and put every result on one page. Compare fill levels, finger counts, poster text, and edits side by side, with the price and generation time under each image.

Today, we are excited to announce that we are joining forces with Stripe, to power the next wave of GDP growth globally.

See what your team spent on every model, save the charts you keep rebuilding, click any bar to land in the logs behind it, and query the same data from your terminal with the Analytics API.

We published live leaderboards grading web search configurations across four task suites. Compare engines, search depth, and models by quality, cost, and speed before your agent makes its first search.

Our new Auto router is informed by the model decisions of millions of people, and it outperforms conventional task-based classifiers across a wide spectrum of tasks.

Using OpenRouter with your favorite harness usually means ad-hoc scripts or a wall of environment variables. Install the ori CLI, log in once, and every supported harness runs with an optimized configuration out of the box.

Choosing the model inside your product is often decided without a systematic method. Ori Eval answers the question with proof: it runs your agent on your own prompts, checks the tools it called, and grades the answers.

Define a taxonomy and a small model tags every generation in your workspace by department, task type, or agent complexity. Filter your logs and group your Activity analytics by the results.

Built from first principles and grounded in timeless Bauhaus philosophy, our new brand identity marks the beginning of our next chapter.

Connect your coding agent to OpenRouter's live model catalog, benchmarks, docs, and test inference, all without leaving your editor.

A dedicated Image API with capability discovery across 60+ models from 12 providers. One endpoint tells your code what each model can do.

The openrouter:subagent server tool lets a frontier model hand off self-contained tasks to a smaller, cheaper, faster worker model mid-generation. Summarize a doc, extract structured data, reformat text, all without burning frontier tokens.

A panel of budget models, fused through OpenRouter, outscored GPT-5.5 and Claude Opus 4.8 on 100 complex research tasks.

The openrouter:advisor server tool lets a fast, cheap model consult a stronger one mid-generation. Run GPT-4o Mini for the routine work. Call Claude Fable when it actually matters.

Speech and transcription APIs, Model Fusion, private models, enterprise workspace controls, and 20 new model launches including Gemini 3.5 Flash and Claude Opus 4.8.

Configurable security and governance tools for budget enforcement, zero data retention, model and provider restrictions, prompt injection defense, and data loss prevention.

OpenRouter has raised a $113M Series B led by CapitalG, with participation from NVentures, ServiceNow Ventures, MongoDB Ventures, Snowflake Ventures, Databricks Ventures, AMP PBC, and Pace Capital, alongside existing investors Andreessen Horowitz and Menlo Ventures.

Give any tool-calling model the ability to search the web and fetch page content on its own, with multiple search engines and fetch engines to choose from.

Text-to-speech and transcription are now live on OpenRouter. Two new endpoints give you access to speech synthesis and audio transcription across multiple providers, under one API.

Introducing the new Response Caching header: enables caching identical API requests so responses come back in a tiny fraction of the time, at zero cost.

Video generation, workspaces, an agent SDK, reranker models, and a wave of frontier model launches.

Run `stripe projects add openrouter/api` to get an OpenRouter account, an API key, and Stripe billing, all from the command line. Your agents can do it too.

Organize your OpenRouter projects into separate environments, each with its own API keys, routing defaults, guardrails, and observability.

Video generation is now live on OpenRouter. One API gives you access to the top video models.

Auto Exacto re-evaluates providers every 5 minutes across throughput, tool-call telemetry, and benchmark scores. For requests that include tools, it's on by default.

Benchmarks on model pages, a new free model router, and improved cost transparency between providers.

What happened and what have we done to prevent this happening again?

Prioritize fast LLMs, explore providers, customize the auto router, load skills in the SDK, and big performance improvements.

Response Healing for structured outputs, chatroom notifications, long-context model rankings, and we're #1 on Brex's fastest-growing AI infra list!

Today we're launching Response Healing, a new feature on OpenRouter that automatically fixes malformed JSON responses from LLMs before they reach your application

Does the same model perform differently across different providers?

An October 2025 update to OpenRouter's “Bring Your Own Key” (BYOK) platform fee.

SOTA image generation, character consistency, and multi-image outputs—now live with Gemini 2.5 Flash Image Preview.

GPT-5 is here on OpenRouter — long-context, built for complex reasoning and code workflows.

Add voice input and send PDFs by URL, on any model.

Venice joins OpenRouter as a new provider, offering its flagship uncensored model for private, powerful, and non-restrictive AI.

Introducing new models to sustain accessible AI inference.

Try Cypher Alpha — a free, general-purpose stealth model with built-in tool-calling.

Centralize your LLM logic, iterate faster, and clean up your code. Presets are now live on OpenRouter.

Track model uptime via API and get more control over your BYOK setup including usage limits and testable keys.