OpenRouter

Step-by-step OpenRouter tutorial for its OpenAI-compatible text-to-speech endpoint: authentication, cURL and Python examples, streaming with the OpenAI SDK, and how to swap models/voices while handling provider-specific format defaults.

Four API-level additions on OpenRouter: GIF prompts accepted by OpenAI, Gemini, Anthropic and Llama models, provider pages in command-K search, caching for Anthropic tool calls, and an is_byok flag in usage-included responses.

Reasoning-model speed metrics on OpenRouter change definition: latency counts time to the first reasoning token and throughput includes reasoning output. A models API category filter and a request builder for generating request JSON also land.

OpenRouter launches a Batch API spanning 70+ models that bundles asynchronous requests at roughly half the standard per-token price, with a median 7-minute turnaround across 230,000+ beta batches.

Gemini joins OpenAI and Anthropic in emitting parallel tool calls through OpenRouter, OAuth PKCE integrations can pass a state parameter through callbacks, provider data-logging policies get their own docs page, and Cent ML and Enfer come online.

OpenRouter launched a hosted, stateful Shell server tool and Files API in beta, letting any model run shell commands in a sandboxed Linux container and read/write files, compatible with OpenAI's and Anthropic's tool specs, billed at $0.0001 per active second.

Eleven models played 30 rounds of a 2D battle royale on $482 of inference. Grok 4.1 Fast took 13 wins at $0.97 each against Claude Sonnet 4.6's five at $26.78, the model with the most kills finished second, and three models won nothing at all.

A field guide to the open-weight models that hold up in real agentic pipelines, with SWE-bench Verified scores, licenses, parameter counts, and realized per-token costs. Includes the argument that the gap to frontier labs has stayed at 3-6 months for a year and a half.

OpenRouter added audio input in wav and mp3 for models that accept it, and direct PDF URL support across every model on the platform, removing the manual base64 encoding step from document workflows.

OpenRouter launches Private Endpoints, an Enterprise add-on that routes fine-tuned models and dedicated deployments through its API, visible only to approved organizations, with BYOK billing and optional fallback to public providers.

OpenRouter identifies its two stealth models as GPT-4.1 and publishes pricing for the generally available 1M context release across the large, Mini and Nano tiers. The free alpha endpoints hard-fail with 404 instead of redirecting to a paid model.

Data residency for teams that consume models through an API is a routing decision, not a datacenter project. Walks the provider-object fields that pin geography and forbid retention or training, how to scope them to the EU, and what happens when no compliant provider can serve the call.

OpenRouter ships a crypto payments API for headless, scripted on-chain top-ups of LLM credit. The practical target is agents that fund their own inference without a person in the billing loop, with a tutorial and docs linked from the post.

OpenRouter measured Opus 4.7's new tokenizer against its own model-agnostic token count using users who switched from 4.6. Native token counts rose 32 to 45 percent depending on prompt size, while real costs rose 12 to 27 percent once caching absorbed part of the inflation.

Compares how LangChain, CrewAI, the OpenAI and Claude agent SDKs, Microsoft Agent Framework and Google ADK handle tool schemas across providers. It also covers what fails on model switches and how API-layer normalization keeps tool calls consistent.

Two OpenRouter skills scaffold a coding agent you own. create-agent-tui builds a terminal harness with configurable tool display, input and banner styles; create-headless-agent builds a structured-IO agent for CLIs, API servers and queue workers. Both sit on the Agent SDK loop.

OpenRouter's February roundup: benchmark scores from Artificial Analysis on every model page, an effective-pricing tab exposing tiered per-provider cost, benchmark scatter charts in rankings, and an openrouter/free router slug for free models.

A cost playbook for OpenRouter: append :floor to route to the cheapest provider for a model, set max_price so a request fails rather than overspends, use :nitro when throughput matters more, and check whether a cheap listing is actually a quantized endpoint.

Investigates whether Seedance 2.5/2.0 models have public weights or a license by checking ByteDance's own pages plus both ByteDance-Seed GitHub and Hugging Face orgs, finds none, and shows the OpenRouter video API call pattern to use instead.

OpenRouter adds streamed reasoning summaries for o3 and o4-mini through its chat completions API, an optional user field for per-customer tracking and moderation, a setting that pins a provider to your own key, crypto invoicing, 2FA and SDK usage accounting.

OpenRouter's live-tested comparison of embedding models by use case: text-embedding-3-small for English RAG, Voyage 4 for long inputs, Qwen3-Embedding for multilingual retrieval, Voyage Code 4 for code search, plus free and low-cost options.

Xiaomi's MiMo-V2.6 family is now live on OpenRouter: a 1T+ parameter flagship (Pro), an open-source MoE (Flash), and a ~10x faster UltraSpeed variant of Pro, all handling text, image, video and audio with 1M context.

Routed-token data showing an open-weight family roughly doubling its share in six months, driven by agentic rather than chat traffic. Includes the price gap against frontier models, adoption split across hobbyist and enterprise keys, and how agentic traffic was classified.

Compares TypeSafe's Jev with LLM-as-a-judge for grading agent outputs, measuring agreement, calibration, bias, latency, and cost, and says which judge fits closed versus open rubrics.