OpenRouter
Named by 49 kept items in this workspace.
39 items name it under frameworks and tools and 10 under classification.technologies.
The items that name it 49
Launch partner platform.
Model routing marketplace; about 165T tokens per week, open-weight models about 72% of tokens.
API access to AI models. Orbio's token originally distributed OpenRouter credits to holders, and Surplus sellers can list OpenRouter access.
Model routing platform whose usage data (tokens, requests, spending) the page analyses; weekly usage approaches 200 trillion tokens.
API that routes requests to any inference provider hosting a model, by default for load balancing, with preferences such as lowest price.
Hosted access; offered free Flint launch access.
Source of the prices used to estimate GB300 profit.
Supported model provider; the default omnara-openrouter provider config uses it.
Model routing platform offering M3.1 Flash free for a limited time.
OpenAI-compatible endpoint usable through a custom base URL.
Model endpoint option that Sphinx can use for the reviewing model.
Site connecting to hundreds of AI models through one account; has a free router sending requests to one of about 20 no-cost models
API gateway that routes requests to different providers hosting the same model, with routing modes Balanced (price + speed), Nitro (fastest) and Floor (cheapest); OpenAI-compatible API with automatic failover and provider routing.
Model routing platform; Step 5 Preview #1 Trending
Inference pricing/routing platform; its inverse-square price routing is said to distort GLM 5.3 output pricing.
Source of undiscounted list-rate token pricing used to compute cost per successful task attempt and cost per dependable task
A service that gives access to hundreds of AI models through one connection; zkAPI's documentation names it as the AI provider behind the short-lived keys, and any app that speaks the OpenAI format can plug in by pointing at a local address.
The model provider the web UI and MCP clients send the conversation and page content to, under the user's API key.
API marketplace/router where Jev is the top model at 1K-10K context
Neutral routing/marketplace layer for LLM inference across many model providers; handles key management, endpoint provisioning, versioning, and discovery so developers don't build this plumbing themselves; reportedly serves 10+ million developers and over 10 trillion tokens per day.
API marketplace used to access multiple models/APIs for agentic builds
Model routing/relay service supported as a magpie provider preset
Platform hosting AssemblyAI Universal-3.5 Pro speech model
A service used to reach the hosted Jev judge, configured via MU_JUDGE_OPENROUTER_API_KEY.
LLM inference platform supplying a single interface to switch between open and closed models from the US and China; publishes usage data showing open model growth from ~1T tokens/week (Sept 2025) to ~80T tokens/week
Model routing provider connection; a plugin can swap the model (not the provider connection), and on OpenRouter that covers every vendor.
Lets you reach Jev through your OpenRouter account and credits
Backend serving Jev over the System One request and response at /api/v1/systemone.
Platform where GLM-5.3-Flash was released as a free-to-use "stealth model" under the name "Ox-Alpha" to build hype before its creator was revealed
Model routing service; armada can route OpenRouter model calls to discounted providers (Novita, StreamLake, Xiaomi) to save up to 20x on cost, with armada models --discounts checking live provider prices and armada init --openrouter-provider setting a preferred provider.
Example vision API provider given as a sample VISION_BASE_URL.
An LLM gateway tool that supports automatic routing based on query complexity.
LLM provider endpoint Graft can use via GRAFT_BASE_URL for its LLM-written concept nodes and summaries
A leading model aggregator and routing layer; its data shows agents have consumed more tokens than humans since February, with agent token usage growing about 14x between February and August versus 2.8x for humans.
A leading model aggregator and routing layer; the page reports that on OpenRouter, agents have consumed more tokens than humans since February, with agent token usage growing about 14x between February and August versus 2.8x for humans.
Supplies the API key (TEXT_MODEL_API_KEY) used to access the text model in the example configuration.
Provides a free model used in the first-agent tutorial
Platform where MiMo-V2.6-Pro-RL is also available
Platform where the model is also available
In every workspace you can read 72
AI
Security
All technologies in this workspace.