OpenRouter

OpenRouter reports that Inception Labs' Mercury 2.5, a diffusion LLM that decodes tokens in parallel, is generally available. It tops OpenRouter's Fastest Models leaderboard among models needing no specialized hardware, with the largest gains on code-heavy prompts. No benchmark figures were publishe

BytePlus's Dreamina Seedance 2.5 is now on OpenRouter as a joint audio-video model. It generates up to 30 seconds per pass, extends across rounds, accepts up to 50 references (30 images, 10 videos, 10 audio), and supports edits such as replacing subjects or adding and removing objects.

OpenRouter's task-level rankings, measured by real spend share rather than benchmark scores, now show DeepSeek V4 Pro leading shell execution and Kimi K3 leading tool dispatch. Both are open-weight models.

OpenRouter published a visual benchmark grid for image models: each model's output on hard prompts such as composing several inputs into one scene, showing a specific number of fingers, or removing the second cup in a row, sortable by price and generation time.

OpenRouter's New Defaults episode with Deedy Das of Menlo Ventures on inference economics: token prices down 40 percent while usage grew fourfold, why optimizing cost too early backfires, and why serious AI products end up spanning multiple models.

Ori now launches Prime Intellect's Agent on OpenRouter with one command, injecting the API key per run through an extension instead of editing auth.json or models.json, and stripping stray Anthropic and OpenAI keys so all requests route through OpenRouter.

OpenRouter shows how to move LLM batches out of web services into Render Workflows, where each prompt is a queued, retried task with on-demand compute. Auto Router records the chosen model per run. It warns that retries can double-bill, so production side effects need idempotency.

Alibaba's Wan 3.0 generates 2 to 30 second videos from text, images, or references at 480p, 720p, or 1080p, priced at $0.05, $0.10, and $0.20 per second with a limited-time 15 percent discount.

Black Forest Labs' FLUX 3 is live on OpenRouter's video API: one jointly trained family spanning video, audio, image, and action prediction, with text-to-video, multi-keyframe image-to-video, video continuation, and native audio including lipsync and ambience.

Microsoft's MAI-Transcribe 2 is now served through OpenRouter. OpenRouter says it ranks first on FLEURS and supports 60 languages, automatic language ID, code switching, speaker diarization and word-level timestamps. The model page was not readable, so details are unverified.

OpenRouter's Activity dashboard gains movement-ranked Trends and an Explore builder over metrics like spend, cache hit rate and P50/P90/P99 latency. A matching Analytics API with a metadata endpoint lets coding agents run cost reviews; one internal review found a $6.2K/month misrouted key.

Grok Imagine Image 2.0 is now on OpenRouter from $0.04 per image. It accepts up to three reference photos per request, plans typography and layout, and supports edits that change one element while leaving the rest of the frame alone. Quality claims come from the announcement only.

Meta Superintelligence Labs released Muse Glimmer, its first open-weights model: a 30B dense text and image model under Apache 2.0, aimed at local agent use. OpenRouter reports 75.5 on MCP Atlas and 51.2 on SWE-Bench Pro, served via zero-retention providers.

OpenRouter updated its :nitro and :floor variants to use service tiers rather than only sorting endpoints. :nitro now includes priority-tier endpoints in the pool alongside defaults, and ~latest resolves to a provider's newest model, so one slug combines all three.

Black Forest Labs' FLUX Video Upscale is available through OpenRouter's video API, upscaling to 2K or 4K in two modes: precise, which is faster and holds identity, and creative, which repairs and generates more detail.

Runway is now served on OpenRouter with Aleph 2.0 and Gen-4.5 behind one API. Aleph 2.0 edits existing footage by changing a single frame and propagating that direction across the video, while Gen-4.5 generates new video with tighter prompt adherence and motion quality.

Voyage Code 4, a code retrieval embedding model from Voyage AI by MongoDB, is now offered on OpenRouter. It supports Matryoshka embeddings at 256, 512, 1024 and 2048 dimensions plus several quantization options. Pricing and benchmarks were not available in the source.

Recraft V4 Styles arrived on OpenRouter in four variants: 1K raster, 2K Pro, and editable SVG versions of each. Requests take 1 to 10 reference images and generate new work matching texture, composition, color and rendering, with no fine-tuning.

Explains Claude's 529 overloaded_error versus a 429 rate limit, and how to keep apps and Claude Code working through capacity shortages with exponential backoff, multi-provider failover, and fallback model lists.

OpenRouter explains how its Fusion compound model works: a panel of models answers in parallel, a judge maps agreement and disagreement, and the calling model synthesizes one answer. Covers the ~4-5x cost multiplier and when the quality gain justifies it.

A unified image generation API across 30+ models adds capability discovery: typed descriptors per model and per provider endpoint covering resolutions, aspect ratios, batch limits, reference-image counts, streaming and the exact billing unit, behind one normalized request shape.

OpenRouter data shows DeepSeek V4.1 Flash handling 1T tokens in its first day and about 2.8T in 48 hours, with 90% cache reads priced near $0.006/M, roughly 5x below GLM-5.3 Flash. Figures cover OpenRouter traffic only.

OpenRouter's new shell server tool and Files API give any model a hosted, isolated Linux container for command execution and persistent files, compatible with OpenAI's Shell and Anthropic's Bash tool specs.

Walkthrough for extracting typed table rows from PDFs with a JSON schema. Covers handling parentheses negatives and footnotes, choosing a parser route for scans, catching truncation, and CSV output, tested on five PDFs.