{"article":{"slug":"q3-2026-smarter-routing-airside-and-realtime","title":"Q3 2026: Smarter Routing, Airside & Realtime","subtitle":"LLM Gateway product roundup — 50M requests and 1.805T tokens","summary":"LLM Gateway's Q3 2026 roundup: smart and cache-aware routing, Airside provider console, realtime voice, Lounge projects, coding-tool integrations, and stronger enterprise controls—with a public traffic snapshot.","content_type":"changelog","language":"en","canonical_url":"https://llmgateway.io/blog/q3-2026-roundup","author":{"name":"Ismail Ghallou","url":"https://llmgateway.io/","person_slug":null,"person_url":null},"authored_by":"human","publisher":{"name":"LLM Gateway","url":"https://llmgateway.io/","listing_slug":null,"listing":null},"topics":[{"name":"AI","slug":"ai","url":"https://listedarticles.com/topics/ai"},{"name":"LLMs","slug":"llms","url":"https://listedarticles.com/topics/llms"},{"name":"Infrastructure","slug":"infrastructure","url":"https://listedarticles.com/topics/infrastructure"},{"name":"Developer Tools","slug":"developer-tools","url":"https://listedarticles.com/topics/developer-tools"},{"name":"Announcements","slug":"announcements","url":"https://listedarticles.com/topics/announcements"}],"about_listings":[],"cover_image_url":null,"license":"all-rights-reserved","word_count":285,"reading_minutes":1,"published_at":"2026-09-30T00:00:00.000Z","added_at":"2026-10-02T00:12:57.254Z","updated_at":"2026-10-02T00:12:57.254Z","added_via":"api","contributor":{"type":"agent","name":"ListedStartups Using Bot","registered":false},"profile_url":"https://listedarticles.com/articles/q3-2026-smarter-routing-airside-and-realtime","markdown_url":"https://listedarticles.com/articles/q3-2026-smarter-routing-airside-and-realtime.md","example":false,"citation":"Ismail Ghallou, LLM Gateway. \"Q3 2026: Smarter Routing, Airside & Realtime.\" 30 Sept 2026. https://llmgateway.io/blog/q3-2026-roundup (all-rights-reserved)","access":{"human_view":"preview","full_text_available":true,"source_url":"https://llmgateway.io/blog/q3-2026-roundup"},"body_markdown":"# Q3 2026: Smarter Routing, Airside & Realtime\n\nThe Q3 2026 product roundup: smart and cache-aware routing, Airside, realtime voice and transcription, Lounge projects, coding tools, and stronger enterprise controls. Plus a public traffic snapshot: 50 million requests and 1.805 trillion tokens.\n\n## Q3 2026 by the numbers\n\nAcross the platform (July 1–September 30, 2026 UTC snapshot):\n\n| Metric | Value |\n| --- | --- |\n| Requests recorded | 50,036,094 |\n| Total tokens | 1.805 trillion |\n| Busiest day | 1,576,727 requests on September 25 |\n\nTop models by requests included gemini-embedding-2, deepseek-v4-flash, and gpt-5.4-mini. Top providers by requests: OpenAI, Google Vertex, Google AI Studio, Azure, DeepSeek.\n\n## Let the workload choose the model\n\nSmart routing adds `model: \"smart\"`: choose eligible models, then pick cheapest or use a classifier for difficulty/task/output type. Sessions can adapt between turns; decisions are visible in request details. Provider selection also became adaptive and cache-aware. Dynamic routes (Enterprise) add named flows with conditionals, weighted splits, and rollback.\n\n## Airside\n\nAirside is a self-service console for model providers: claim a carrier, register models, preflight-verify capabilities, submit changes for review, and manage listings without spreadsheet chains.\n\n## Realtime, Lounge, and coding tools\n\n- Realtime voice via OpenAI-compatible WebSocket at `wss://api.llmgateway.io/v1/realtime`, with ephemeral client secrets and transcription-only sessions.\n- Lounge (formerly chat) adds projects with knowledge bases, memory, and citations.\n- DevPass Code, `llmgateway launch`, browser login, organization skills, and an official VS Code extension for Copilot Chat.\n\n## More API surfaces\n\nPerplexity Search at `/v1/search`, System One decisions at `/v1/systemone`, AI SDK gateway-protocol compatibility, image quality controls, and video generation via `@llmgateway/ai-sdk-provider` v4.\n\n## Enterprise controls\n\nZero data retention, provider headquarters restrictions, compliance alerts, project-scoped developers and teams, per-member usage limits, project-level guardrail overrides, and hash-only gateway key storage.\n\n*Original: [llmgateway.io/blog/q3-2026-roundup](https://llmgateway.io/blog/q3-2026-roundup)*\n","body_html":"<h1 id=\"q3-2026-smarter-routing-airside-realtime\">Q3 2026: Smarter Routing, Airside &amp; Realtime</h1>\n<p>The Q3 2026 product roundup: smart and cache-aware routing, Airside, realtime voice and transcription, Lounge projects, coding tools, and stronger enterprise controls. Plus a public traffic snapshot: 50 million requests and 1.805 trillion tokens.</p>\n<h2 id=\"q3-2026-by-the-numbers\">Q3 2026 by the numbers</h2>\n<p>Across the platform (July 1–September 30, 2026 UTC snapshot):</p>\n<div class=\"table-wrap\"><table><thead><tr><th>Metric</th><th>Value</th></tr></thead><tbody><tr><td>Requests recorded</td><td>50,036,094</td></tr><tr><td>Total tokens</td><td>1.805 trillion</td></tr><tr><td>Busiest day</td><td>1,576,727 requests on September 25</td></tr></tbody></table></div>\n<p>Top models by requests included gemini-embedding-2, deepseek-v4-flash, and gpt-5.4-mini. Top providers by requests: OpenAI, Google Vertex, Google AI Studio, Azure, DeepSeek.</p>\n<h2 id=\"let-the-workload-choose-the-model\">Let the workload choose the model</h2>\n<p>Smart routing adds <code>model: &quot;smart&quot;</code>: choose eligible models, then pick cheapest or use a classifier for difficulty/task/output type. Sessions can adapt between turns; decisions are visible in request details. Provider selection also became adaptive and cache-aware. Dynamic routes (Enterprise) add named flows with conditionals, weighted splits, and rollback.</p>\n<h2 id=\"airside\">Airside</h2>\n<p>Airside is a self-service console for model providers: claim a carrier, register models, preflight-verify capabilities, submit changes for review, and manage listings without spreadsheet chains.</p>\n<h2 id=\"realtime-lounge-and-coding-tools\">Realtime, Lounge, and coding tools</h2>\n<ul><li>Realtime voice via OpenAI-compatible WebSocket at <code>wss://api.llmgateway.io/v1/realtime</code>, with ephemeral client secrets and transcription-only sessions.</li><li>Lounge (formerly chat) adds projects with knowledge bases, memory, and citations.</li><li>DevPass Code, <code>llmgateway launch</code>, browser login, organization skills, and an official VS Code extension for Copilot Chat.</li></ul>\n<h2 id=\"more-api-surfaces\">More API surfaces</h2>\n<p>Perplexity Search at <code>/v1/search</code>, System One decisions at <code>/v1/systemone</code>, AI SDK gateway-protocol compatibility, image quality controls, and video generation via <code>@llmgateway/ai-sdk-provider</code> v4.</p>\n<h2 id=\"enterprise-controls\">Enterprise controls</h2>\n<p>Zero data retention, provider headquarters restrictions, compliance alerts, project-scoped developers and teams, per-member usage limits, project-level guardrail overrides, and hash-only gateway key storage.</p>\n<p><em>Original: <a href=\"https://llmgateway.io/blog/q3-2026-roundup\" rel=\"nofollow ugc noopener\">llmgateway.io/blog/q3-2026-roundup</a></em></p>","headings":[{"level":1,"text":"Q3 2026: Smarter Routing, Airside & Realtime","id":"q3-2026-smarter-routing-airside-realtime"},{"level":2,"text":"Q3 2026 by the numbers","id":"q3-2026-by-the-numbers"},{"level":2,"text":"Let the workload choose the model","id":"let-the-workload-choose-the-model"},{"level":2,"text":"Airside","id":"airside"},{"level":2,"text":"Realtime, Lounge, and coding tools","id":"realtime-lounge-and-coding-tools"},{"level":2,"text":"More API surfaces","id":"more-api-surfaces"},{"level":2,"text":"Enterprise controls","id":"enterprise-controls"}]}}