Latest news and updates from LLM Gateway
Automatic model selection now reads the request before it routes it. Choose the models `smart` may resolve to, and a classifier rates each prompt so a one-line fix goes to a cheap model and a hard problem goes to a frontier one. Available to every organization while in beta.
A worked model of LLM cost optimization for a team spending $100k a month: what difficulty-based model routing saves at three traffic mixes, what the classifier costs, and the break-even point. Real catalogue prices, explicit assumptions, and the arithmetic to run on your own numbers.
Compliance alerts watch the models your provider compliance policy blocks today and notify your team the moment a provider that meets your requirements starts serving one — in-app, by email, or in Slack. Downgrade alerts fire when a provider stops meeting the policy. Available on the Enterprise plan.
The two-rappers-swapped clips flooding your feed are one image edit plus one image-to-video call. This guide recreates the trend with Steve Jobs and Bill Gates on MiniMax H3 Max through the LLM Gateway API and the Lounge Video Studio, for about $1.40.
The SCX model discount was due to end on September 24. It now runs 15 more days: 20% off the same seven models on LLM Gateway until October 9, 2026. Nothing changes in how you route requests.
Most classification and routing calls use a chat model, then parse prose and hope the JSON holds. The new /v1/systemone LLM decision API on LLM Gateway returns typed yes/no, choice, and score answers with calibrated probabilities, billed on input tokens only.
Perplexity retires its Sonar chat completions API on September 27, 2026. On LLM Gateway, perplexity/sonar moves to Perplexity's Agent API on September 25 with no changes to your requests and lower prices, while sonar-pro and sonar-reasoning-pro stop being routable on September 27. Here is exactly what changes, why, and what to do about it.
We gave six models the same coding task and measured what each one actually cost. The cheapest produced a working result for 10 cents. The most expensive spent $198 on a single index.html. This AI coding model benchmark shows where the money goes and where it stops buying quality.
A checklist for inference providers who want their LLM API listed on an AI gateway: domain ownership, an OpenAI-compatible endpoint the gateway can actually call, honest capability flags, streaming that reports usage, flat per-million pricing, and rate limits you can sustain. Written against LLM Gateway's Airside listing process.
LLM routing on LLM Gateway is an election: every provider listing a model is scored on effective price, uptime, throughput, and latency, and the lowest score wins. This guide explains the scoring from the provider's side, with the fare formula, the uptime penalty, cache-aware pricing, and the levers a carrier controls from Airside.
A practical guide to making money with LLM inference: list your deployment on LLM Gateway through Airside, file the prices developers pay, choose the landing fee you accept, and keep the rest of the traffic you win. Includes the fare math, a worked example, and the mistakes that cost providers revenue.
The SCX model discount brings 20% off seven selected models on LLM Gateway, starting when the Runware promotion ends on September 9 and now extended through October 9. See the eligible models and how to route requests through SCX.
AI SDK video generation now works through LLM Gateway: version 4 of @llmgateway/ai-sdk-provider targets AI SDK 7 and adds llmgateway.video() for experimental_generateVideo. Generate Veo, Seedance, and KLING clips with one API key, including image-to-video, first and last frames, references, and long-running jobs.
Cache-aware LLM routing learns cache-hit rates and token proportions from your project's recent model usage, so provider selection reflects the workload you actually run. Available automatically across plans, with explicit per-project overrides on Enterprise.
The LLM Gateway CLI now signs in through your browser or enterprise SSO, installs your organization's shared skills into Claude Code, Codex, OpenCode, Cursor, and more, and launches 15 coding agents pre-wired to the gateway. Organization skills are an Enterprise feature; browser login works on every plan.
Airside is the self-serve carrier console where LLM providers claim their listing on LLM Gateway, register models, file prices for review, and tune the discount and margin that win routed traffic. Listing costs a one-time $2,500 fee per provider company, and every model is verified live before it goes on the departure board, with a per-request breakdown of what your endpoint served.
A step-by-step guide to LLM usage and cost tracking without leaving the terminal: connect the LLM Gateway MCP server to Claude Code, Codex, or Cursor, then ask for spend, token totals, cache hits, and rankings by model, provider, app, or API key. The tools are read-only and free to call.
The best Cursor alternatives in 2026, compared honestly — AI editors, terminal agents, and gateway-backed setups. Where Cursor's credit pools and curated model list pinch, and which alternative fixes each complaint without breaking your workflow.
GLM-5.2 in Claude Code takes three environment variables. This guide walks through the exact setup for Claude Code, Cursor, and Cline via LLM Gateway — plus why GLM-5.2 is the rare open coder that's uncapped on a flat-rate DevPass plan.
OpenCode Go replaced its request-count tiers with dollar metering: $10/month now buys up to $60 of usage at published per-token rates, with Grok 4.5 and GPT-5.6 Luna capped at $15 each. Here's exactly what changed, the math per dollar, and when the ceiling stops fitting.
TanStack AI now ships a first-party LLM Gateway adapter. Install @tanstack/ai-llmgateway, add one API key, and your useChat app can stream chat, call tools, and surface reasoning from 200+ models across 40+ providers — switching models is a one-line change.
Microsoft splits the two big AI use cases across separate products: Azure AI Foundry for calling models over an API, GitHub Copilot for coding agents — plus a new API Management gateway tier if you want governance. Here is what each one costs in 2026, and what the same two workloads cost on a single gateway.
Stripe and OpenRouter confirmed the deal on August 19, 2026, three days after Bloomberg reported it. Here's what is verified, what OpenRouter says changes for customers (nothing, for now), and how to check whether you could leave any gateway if you needed to.
Your coding agent started erroring and you need to know whether Kimi is down or your request is wrong. Here is how to read Kimi K3 API status per provider, tell an outage from a client error, and keep working through both.
An AI coding plan is a flat monthly fee that covers model usage inside your coding agent. This explains what a coding plan actually includes, how monthly allowances and weekly premium caps work, and how to size one against your real usage.
Switching off the Vercel AI Gateway used to mean rewriting how your AI SDK app resolves models — and losing provider-native web search on the way out. LLM Gateway now implements the AI SDK's own gateway protocol, so the migration is one line and your bare model strings keep working.
In the August 7 run of computesdk's independent AI gateway benchmark, LLM Gateway ranked first of six gateways with a composite score of 90.8 — driven by the tightest tail latency in the field, not the fastest median. Here's what the benchmark measures, what our numbers actually show, and how to run it yourself.
How to keep LLM traffic GDPR-compliant in practice: restrict routing to GDPR-compliant providers, filter by provider headquarters, pin regions, and block non-compliant requests before any data leaves your gateway.
Helicone entered maintenance mode after the Mintlify acquisition in March 2026. The 8 best Helicone alternatives in 2026, compared honestly — gateways with built-in analytics and dedicated LLM observability platforms — and how to pick between them.
How to use the Kimi K3 API in practice: endpoints, pricing ($3/$0.30/$15 per million tokens), the reasoning_effort parameter, tool calls, JSON output, the 1M-token context — and how to route it through one OpenAI-compatible endpoint with automatic failover.
Moonshot released the Kimi K3 open weights on July 26, 2026 — 2.8T parameters, the largest open-weight model ever shipped. What's in the release, what it takes to run Kimi K3 locally, the license caveat, and the hosted routes that don't need a GPU cluster.
An eight-point LLM compliance checklist for putting model traffic in production: provider vetting, routing restrictions, data residency, retention, access control, and the audit trail — with what to enforce at the gateway versus document by hand.
A practical guide to LLM data retention: when storing full prompts and responses helps, when metadata is enough, what it costs, and how to set a retention policy your compliance team will sign off on.
The 8 best Portkey alternatives in 2026, compared honestly — open-source gateways, managed routers, and VPC platforms — with fees, self-hosting, governance, and who each option genuinely fits after the Palo Alto Networks acquisition.
A text-to-speech API tutorial: synthesize speech with ElevenLabs, OpenAI, Gemini, and Qwen voices through one OpenAI-compatible endpoint, choose voices and formats, steer delivery with instructions, and track the cost of every clip.
A practical AI image generation API tutorial: call gpt-image-2, Gemini, and Seedream through one OpenAI-compatible endpoint, pick sizes and quality, edit existing images, and see the exact cost of every generation.
An AI video generation API tutorial: submit async jobs to Veo, Seedance, and KLING through one OpenAI-compatible endpoint, poll or receive signed webhooks, download the MP4, and keep per-second costs visible.
An LLM cost tracking tutorial: read the exact USD cost of every request from the response's usage object, segment spend by user and feature with metadata headers, enforce hard budgets with per-key spending limits, and audit everything in the dashboard.
The best open-source OpenRouter alternatives in 2026 — self-hostable AI gateways compared by what ships in the open core, what stays paid, and what you'll actually operate.
Six enterprise OpenRouter alternatives compared on the criteria procurement actually checks — self-hosting and VPC deployment, SSO and audit logs, guardrails, SLAs, and compliance posture.
The 10 best OpenRouter alternatives in 2026, compared honestly — open-source gateways, managed routers, and enterprise platforms — with fees, self-hosting, and routing side by side.
We measured AI gateway performance with an open-source TTFT benchmark: 75 cold + 75 warm interleaved runs of claude-haiku-4.5 against LLM Gateway and OpenRouter, with phase-by-phase timings and raw data published. LLM Gateway reached first token ~35% faster cold and ~34% faster warm, and here is exactly how to reproduce the numbers.
The best open-source LLMs in 2026, ranked — Kimi K3, GLM-5.2, DeepSeek V4 Pro, MiniMax M3 and more, compared on license, context window, and real per-token price. All of them run through one API with LLM Gateway.
Kimi K3 in Claude Code takes three environment variables. This guide walks through the exact setup for Claude Code, Cursor, and Cline via LLM Gateway — plus what each tool does and doesn't route, and what K3 costs on a flat-rate DevPass plan.
Kimi K3 ties Claude Opus 4.8 on GPQA Diamond, costs 40% less per token, and its weights are public while Opus stays closed — but Opus still leads where it counts for some teams. A fact-checked comparison of benchmarks, pricing, and context windows, and how to A/B both through one API.
Kimi K3 is the largest open-weight model ever released — 2.8T parameters, a 1M-token context, and benchmark scores next to Claude Opus 4.8. Here is what it costs, how it compares to GLM-5.2, DeepSeek V4 Pro, and MiniMax M3, and how to run all of them through one API with LLM Gateway — on a flat-rate DevPass plan or pay-as-you-go credits.
GitHub Copilot switched chat and agents to usage-based AI Credits on June 1, 2026, and bills jumped 10–50x for agentic teams. The best GitHub Copilot alternatives in 2026, compared honestly — flat-fee IDEs, open-source agents, and gateway-backed setups with hard spend caps.
Microsoft repriced enterprise Copilot three times in June 2026: GitHub Copilot moved to usage-based AI Credits, Copilot Cowork added per-task billing on top of the $30 seat, and volume discounts expired. What it costs now — and the cost-control playbook enterprises are adopting instead.
The best LiteLLM alternatives in 2026, compared honestly — open-source gateways, managed routers, and enterprise platforms — and how to pick the right one.
LLM Gateway Chat now has Projects: group related chats, upload files — PDFs and spreadsheets included — as a knowledge base, and get answers grounded in your own documents via RAG, with source citations, on any of 280+ models. Projects also remember durable facts across chats. Available to every Chat user.
DevPass Code is an open-source terminal coding agent that talks only to LLM Gateway. One browser login, every model, and no per-provider API keys to juggle. Use it pay-as-you-go or on a DevPass coding subscription.
Three months of updates: speech generation and Audio Studio, OpenAI-compatible embeddings, OCR, DevPass coding plans, chat subscription plans, enterprise IAM and master keys, SOC 2 Type II, 40+ new models, and much more.
Most dashboards show what you spent, not where it went. LLM Gateway's enterprise LLM analytics break cost, requests, and tokens down by model, project, API key, and team member — no data warehouse to build. Member and organization-wide analytics are available on the Enterprise plan.
AI gateway pricing hides in two places — the platform fee on credits and the markup on tokens. Here's what OpenRouter, Vercel, Cloudflare, Portkey, LiteLLM, and LLM Gateway actually charge, and how to pay the least.
Claude Opus 4.8 lists at $5/$25 per million tokens — the same as Opus 4.6. But Anthropic's newer tokenizer can turn the same text into up to ~35% more tokens, so your real bill can climb even when the sticker price doesn't.
An honest comparison of the best AI coding plans in 2026 — Claude Code, Cursor, Copilot, Codex, OpenCode Go and more — ranked on price, model access, and lock-in. DevPass tops the list with one flat rate for every model.
The best ChatGPT alternatives in 2026, compared honestly on models, features and price. Lounge by LLM Gateway leads — every frontier model plus image generation on one membership from $9/mo.
What it takes to run LLM Gateway in production on AWS, GCP, or Azure — the components you need, how they fit together, and why Kubernetes is the path we recommend once you outgrow a single box.
Rotating API keys shouldn't cause service interruption for production AI. Learn how LLM Gateway enables secure key rotation for both providers and the gateway.
A walkthrough of our new open-source template: a Slack bot that streams AI answers, keeps thread context, and searches the web — backed by LLM Gateway so you can switch between 200+ models with one API key.
LLM Gateway has successfully completed its SOC 2 Type II audit. Here's what that means, why it matters for teams routing LLM traffic through us, and how to request our report.
Nine text-to-speech models from ElevenLabs, OpenAI, and Google are now one OpenAI-compatible API call away — plus a new Audio Studio in the Playground to compare voices and models side by side.
Our new Payments SDK lets your end-users buy credits inside your app and pay per request for any model — billed through LLM Gateway, with your markup as margin. Here's how it works and how to ship it in ~40 lines.
Looking for a Portkey alternative? A straightforward comparison of LLM Gateway and Portkey — features, pricing, deployment, and trade-offs — so you can pick the right AI gateway for your stack.
LLM orchestration is the layer that coordinates models, providers, and steps into one reliable workflow. A practical guide to the patterns, the tools, and when you need an LLM orchestrator.
All three ByteDance Seedance video generation models — 2.0, 2.0 Fast, and 1.5 Pro with native audio — are now live on LLM Gateway. Same API key. Same billing. Same dashboard.
Generate vectors for semantic search, clustering, and RAG through the same gateway you already use for chat. OpenAI-compatible, drop-in, and tracked alongside your model spend.
Calling OpenAI or Anthropic directly is the right first call. Here's the honest case for when a gateway starts paying for itself — and when you don't need one yet.
How LLM response caching actually works, where it helps, where it doesn't, and how to turn it on without rewriting your app.
A practical guide to forecasting LLM costs: the token formula, real-world examples across GPT-5.4, Claude, and Gemini, and a free calculator to run the numbers.
A straightforward comparison of LLM Gateway and LiteLLM — features, operational cost, and trade-offs — so you can pick the right one for your stack.
No config files, no env vars. OpenCode ships LLM Gateway as a first-class provider — select it, paste your key, and start coding with 200+ models.
A practical framework for picking the right model — based on task type, budget, latency requirements, and context window — instead of chasing benchmarks.
A deep dive into the routing, retry, and failover systems that keep LLM Gateway reliable when upstream providers go down.
A straightforward comparison of LLM Gateway and OpenRouter — features, pricing, and trade-offs — so you can pick the right one for your stack.
What LLM guardrails are, why they matter in production, and how to implement content safety without building it yourself.
An honest comparison of the top AI gateways — features, pricing, and trade-offs — so you can pick the right one for your stack.
LLM Gateway offers exclusive discounts on DeepSeek V3.2, V3.1, and R1 through partner providers — up to 30% off base pricing, applied automatically to every request.
We compared DeepSeek V3.2 pricing across every major API provider. Here's the definitive ranking — and how our Token Cost Calculator can help you estimate exact savings.
Three months of updates: video generation, Image Studio, sessions, GPT-5.4 family, enterprise guardrails, 5+ new providers, and much more.
A practical breakdown of how intelligent routing, caching, and model selection through an LLM gateway can dramatically reduce your AI infrastructure costs.
Side-by-side pricing comparison of GPT-5, Claude Opus 4.8, and Gemini 2.5 Pro with real cost calculations for production workloads.
LLM Gateway partners with Alibaba Cloud to bring you 20% off 26 Qwen AI models — including Qwen3 Max, Qwen3 Coder, QwQ reasoning, vision-language, and image generation models.
A step-by-step guide to making your first LLM API request through LLM Gateway — from signup to seeing results in your dashboard.
Why building directly against a single LLM provider's API is riskier than you think, and how a gateway layer protects your AI investment.
What an LLM gateway does, why it matters, and how it lets you ship AI features faster by abstracting away provider complexity.
Learn why simple LLM proxies aren't enough and how a unified AI gateway delivers centralized access control, cost visibility, compliance, and security.
Learn what an LLM Gateway is, why you need one, and how it simplifies integrating, managing, and deploying large language models in production.
How we deploy our Next.js apps on Google Cloud Platform without relying on Vercel.
Three months of updates: 15+ new models, team management, referral program, tiered pricing, data retention, new providers, and much more.
Use the CYBERMONDAY promo code to get 50% off credits for a limited time.
Use the BLACKFRIDAY promo code to get 50% off credits for a limited time.
Compare GPT-5, Claude, Gemini, and 280+ other models side by side. Generate images, test prompts, and find the best model for your use case.
Use GPT-5, Gemini, or any model with Claude Code. Three environment variables, zero code changes.
Connect your internal LLM deployments or any OpenAI-compatible API to LLM Gateway—and get the same analytics, caching, and routing.
Run LLM Gateway on your own infrastructure in under 5 minutes. Full control, zero platform fees.
One API for 280+ models across 35+ providers. Route requests, track costs, and switch models without changing your code.