OpenRouter review
Updated September 28, 2026 · DeployCue Editorial Team · Facts checked September 26, 2026
We earn a commission when you sign up through links on this site. Partner status can affect where a brand appears in our rankings. Partners with a live offer in your country come first, then every other brand by score.
Our verdict
OpenRouter suits teams that want one OpenAI-compatible endpoint across many models and providers without hosting a gateway. It routes requests across from 80 providers, holds SOC 2 Type 2 certification and charges pay as you go per token plus a platform fee on credit purchases, with a free tier.
Key facts
- Best for
- One key and one bill for teams routing across many models and providers.
- Fee model
- Pay-as-you-go per token plus a platform fee on credit purchases: 5.5% (Standard), 8% (Business), fee discounts (Enterprise); free tier has no platform fee↗
- Providers routed
- from 80↗
- Bring your own key
- Yes↗
- OpenAI-compatible API
- Yes↗
- Free tier
- Yes↗
- Compliance certifications
- SOC 2 Type 2↗
- Legal entity
- OpenRouter, Inc↗
- Headquarters
- United States↗
- Founded
- 2023↗
- Support channels
- Email, Help-centre, Community↗
- Official website
- openrouter.ai
LLM APIs
- Input price from
- $0↗
- Output price from
- $0↗
- Models offered
- Respan/span-01, Respan/span-01-lite, Respan/span-01-lite:free, Bytedance-seed/seed-audio-1-0, Typesafe/jev-router, Jaredpalmer/kev-4b, Perceptron/perceptron-mk1.5, Google/gemini-3.5-transcribe, Fish-audio/transcribe-1-pro, Fireworks/ember-1, Inclusionai/ming-image-0.1-design-layer, Google/gemini-3.8-flash-lite-tts, Google/gemini-3.8-flash-tts, Z-AI/glm-5.3-prime, Qwen/qwen3.8-max-prime, Recraft/recraft-v4.1-flash, Stealth/space-bunny-alpha, Aion-labs/aion-3.5-mini, Aion-labs/aion-3.5, Upstage/solar-mini4↗
- Largest context window
- 1,050,000↗
- Free credits
- Small free allowance for new users; 50 free-model requests per day (1,000 per day after buying at least $10 credits)↗
Facts checked September 26, 2026, 7 sources
Apps and profiles
Pricing
Qwen3 Next 80B A3B via OpenRouter (USD)
input-tokens: $0.09 / million
Qwen3 Next 80B A3B via OpenRouter (USD)
output-tokens: $1.10 / million
Qwen3 Coder 30B A3B via OpenRouter (USD)
input-tokens: $0.30 / million
Qwen3 Coder 30B A3B via OpenRouter (USD)
output-tokens: $1.00 / million
Qwen3 32B via OpenRouter (USD)
input-tokens: $0.08 / million
Qwen3 32B via OpenRouter (USD)
output-tokens: $0.28 / million
Qwen3 30B A3B via OpenRouter (USD)
input-tokens: $0.09 / million
Qwen3 30B A3B via OpenRouter (USD)
output-tokens: $0.30 / million
Qwen3 235B A22B via OpenRouter (USD)
input-tokens: $0.09 / million
Qwen3 235B A22B via OpenRouter (USD)
output-tokens: $0.55 / million
Qwen 2.5 Coder 32B via OpenRouter (USD)
input-tokens: $0.66 / million
Qwen 2.5 Coder 32B via OpenRouter (USD)
output-tokens: $1.00 / million
Qwen 2.5 7B via OpenRouter (USD)
input-tokens: $0.10 / million
Qwen 2.5 7B via OpenRouter (USD)
output-tokens: $0.20 / million
Qwen 2.5 72B via OpenRouter (USD)
input-tokens: $0.36 / million
Qwen 2.5 72B via OpenRouter (USD)
output-tokens: $0.40 / million
Phi-4 via OpenRouter (USD)
input-tokens: $0.07 / million
Phi-4 via OpenRouter (USD)
output-tokens: $0.14 / million
OpenAI o4-mini via OpenRouter (USD)
input-tokens: $1.10 / million
OpenAI o4-mini via OpenRouter (USD)
output-tokens: $4.40 / million
OpenAI o3-mini via OpenRouter (USD)
input-tokens: $1.10 / million
OpenAI o3-mini via OpenRouter (USD)
output-tokens: $4.40 / million
OpenAI o3 via OpenRouter (USD)
input-tokens: $2.00 / million
OpenAI o3 via OpenRouter (USD)
output-tokens: $8.00 / million
OpenAI o1 via OpenRouter (USD)
input-tokens: $15.00 / million
OpenAI o1 via OpenRouter (USD)
output-tokens: $60.00 / million
Mixtral 8x22B via OpenRouter (USD)
input-tokens: $2.00 / million
Mixtral 8x22B via OpenRouter (USD)
output-tokens: $6.00 / million
Mistral NeMo via OpenRouter (USD)
input-tokens: $0.03 / million
Mistral NeMo via OpenRouter (USD)
output-tokens: $0.03 / million
Mistral Medium 3 via OpenRouter (USD)
input-tokens: $0.40 / million
Mistral Medium 3 via OpenRouter (USD)
output-tokens: $2.00 / million
MiniMax M2 via OpenRouter (USD)
input-tokens: $0.26 / million
MiniMax M2 via OpenRouter (USD)
output-tokens: $1.02 / million
Llama 4 Scout via OpenRouter (USD)
input-tokens: $0.10 / million
Llama 4 Scout via OpenRouter (USD)
output-tokens: $0.30 / million
Llama 4 Maverick via OpenRouter (USD)
input-tokens: $0.19 / million
Llama 4 Maverick via OpenRouter (USD)
output-tokens: $0.65 / million
Llama 3.3 70B via OpenRouter (USD)
input-tokens: $0.10 / million
Llama 3.3 70B via OpenRouter (USD)
output-tokens: $0.32 / million
Llama 3.1 8B via OpenRouter (USD)
input-tokens: $0.05 / million
Llama 3.1 8B via OpenRouter (USD)
output-tokens: $0.08 / million
Kimi K2 via OpenRouter (USD)
input-tokens: $0.57 / million
Kimi K2 via OpenRouter (USD)
output-tokens: $2.30 / million
gpt-oss-20b via OpenRouter (USD)
input-tokens: $0.02 / million
gpt-oss-20b via OpenRouter (USD)
output-tokens: $0.09 / million
gpt-oss-120b via OpenRouter (USD)
input-tokens: $0.15 / million
gpt-oss-120b via OpenRouter (USD)
output-tokens: $0.60 / million
GPT-5 nano via OpenRouter (USD)
input-tokens: $0.05 / million
GPT-5 nano via OpenRouter (USD)
output-tokens: $0.40 / million
GPT-5 mini via OpenRouter (USD)
input-tokens: $0.25 / million
GPT-5 mini via OpenRouter (USD)
output-tokens: $2.00 / million
GPT-5 via OpenRouter (USD)
input-tokens: $1.25 / million
GPT-5 via OpenRouter (USD)
output-tokens: $10.00 / million
GPT-4o mini via OpenRouter (USD)
input-tokens: $0.15 / million
GPT-4o mini via OpenRouter (USD)
output-tokens: $0.60 / million
GPT-4o via OpenRouter (USD)
input-tokens: $2.50 / million
GPT-4o via OpenRouter (USD)
output-tokens: $10.00 / million
GPT-4.1 nano via OpenRouter (USD)
input-tokens: $0.10 / million
GPT-4.1 nano via OpenRouter (USD)
output-tokens: $0.40 / million
GPT-4.1 mini via OpenRouter (USD)
input-tokens: $0.40 / million
GPT-4.1 mini via OpenRouter (USD)
output-tokens: $1.60 / million
GPT-4.1 via OpenRouter (USD)
input-tokens: $2.00 / million
GPT-4.1 via OpenRouter (USD)
output-tokens: $8.00 / million
GLM-4.6 via OpenRouter (USD)
input-tokens: $0.30 / million
GLM-4.6 via OpenRouter (USD)
output-tokens: $0.90 / million
GLM-4.5 via OpenRouter (USD)
input-tokens: $0.13 / million
GLM-4.5 via OpenRouter (USD)
output-tokens: $0.85 / million
Gemma 3 27B via OpenRouter (USD)
input-tokens: $0.08 / million
Gemma 3 27B via OpenRouter (USD)
output-tokens: $0.35 / million
Gemma 2 27B via OpenRouter (USD)
input-tokens: $0.65 / million
Gemma 2 27B via OpenRouter (USD)
output-tokens: $0.65 / million
Gemini 2.5 Pro via OpenRouter (USD)
input-tokens: $1.25 / million
Gemini 2.5 Pro via OpenRouter (USD)
output-tokens: $10.00 / million
Gemini 2.5 Flash via OpenRouter (USD)
input-tokens: $0.30 / million
Gemini 2.5 Flash via OpenRouter (USD)
output-tokens: $2.50 / million
DeepSeek V3.1 via OpenRouter (USD)
input-tokens: $0.27 / million
DeepSeek V3.1 via OpenRouter (USD)
output-tokens: $1.00 / million
DeepSeek R1 Distill via OpenRouter (USD)
input-tokens: $0.80 / million
DeepSeek R1 Distill via OpenRouter (USD)
output-tokens: $0.80 / million
DeepSeek R1 via OpenRouter (USD)
input-tokens: $0.50 / million
DeepSeek R1 via OpenRouter (USD)
output-tokens: $2.15 / million
Command A via OpenRouter (USD)
input-tokens: $2.50 / million
Command A via OpenRouter (USD)
output-tokens: $10.00 / million
Claude Sonnet 4.6 via OpenRouter (USD)
input-tokens: $3.00 / million
Claude Sonnet 4.6 via OpenRouter (USD)
output-tokens: $15.00 / million
Claude Opus 4.8 via OpenRouter (USD)
input-tokens: $5.00 / million
Claude Opus 4.8 via OpenRouter (USD)
output-tokens: $25.00 / million
Claude Haiku 4.5 via OpenRouter (USD)
input-tokens: $1.00 / million
Claude Haiku 4.5 via OpenRouter (USD)
output-tokens: $5.00 / million
Pros and cons
Pros
- Routes requests across from 80 providers behind a single OpenAI-compatible API.
- Free tier and free credits for new users, with a daily cap of free model requests that rises once credits are bought.
- SOC 2 Type 2 certification for teams that need vendor assurance.
- Model pricing starts at $0 for input and $0 for output.
Cons
- Costs depend on which models you call, since pricing is pay as you go per token rather than a flat plan rate.
- Support runs on email, help centre and community, with no phone or live chat line listed.
- The platform fee on credit purchases is higher on Business than on Standard.
Who it suits
- Teams that want many models and providers behind a single OpenAI-compatible API.
- Builders testing ideas on a free tier and free credits before spending on tokens.
- Buyers who need SOC 2 Type 2 certification from a US entity, OpenRouter, Inc.
Who should look elsewhere
- LiteLLM, if you would rather self host an open source gateway and skip the platform fee on credit purchases.
- Salad, if a flat monthly subscription or pay per token fits your billing better than per model rates.
- UCloud, if a provider set of from 4 covers what you need rather than from 80.
How OpenRouter compares
OpenRouter vs WaveSpeedAI9.0 vs 8.8
OpenRouter vs ScalaHosting9.0 vs 8.3
OpenRouter vs Contabo9.0 vs 9.2
OpenRouter vs Cloudways9.0 vs 9.2
OpenRouter vs PlanetHoster9.0 vs 8.7
OpenRouter vs Acronis9.0 vs 9.9
OpenRouter vs Aruba9.0 vs 9.3
OpenRouter vs Decodo9.0 vs 9.1
OpenRouter vs Hostinger9.0 vs 9.1- OpenRouter vs MarsProxies9.0 vs 9.0
OpenRouter vs hide.me9.0 vs 8.7
OpenRouter vs Proton Drive9.0 vs 8.3
OpenRouter vs Nexos.ai9.0 vs 8.2
Frequently asked questions
How much does OpenRouter cost?+
Pricing is pay as you go per token, plus a platform fee on credit purchases that is higher on Business than on Standard and discounted on Enterprise. Model rates start from $0 for input and $0 for output, so free model variants sit at the cheap end of the catalogue.
Is there a free tier?+
There is a free tier, and the platform fee does not apply to it. New users also receive a small free credit allowance, with a set number of free model requests each day and a larger daily cap once credits have been bought.
Which models and providers can I reach through one endpoint?+
The gateway routes across from 80 providers, and the catalogue covers text, image, audio and transcription work, including free model variants. That breadth lets a team move between models without rewriting its integration, as long as calls keep to the OpenAI-compatible shape.
Can I bring my own API keys?+
Yes. Bring your own key is supported, so calls can run on keys you already hold. Routing still spans from 80 providers, and the platform fee on credit purchases still applies when you top up credits.
Is the API OpenAI-compatible?+
Yes, an OpenAI-compatible API is supported, which is the practical reason teams pick this gateway: existing SDK calls and integrations can be repointed at a single base URL. Listed rates start from $0 for input and $0 for output.
How large a context window do I get?+
The largest context window across the catalogue is 1,050,000, set by the widest models listed. Context length comes from the model rather than from a plan, so the practical ceiling is whatever the model you call allows.
How do I get help?+
Support runs on email, help centre and community, so help comes by email, a help centre and the community rather than a phone or live chat line. Since billing follows token volume, a question is easier to settle with the model name and request details to hand.
Who runs OpenRouter and is it certified?+
The service runs as OpenRouter, Inc, headquartered in the US as given by US and founded in 2023. Assurance is published as SOC 2 Type 2 certification, which matters for buyers who want vendor paperwork before routing production traffic.
Sources
- openrouter.ai/about checked September 26, 2026
Backs: Founded, Providers routed
“Started in early 2023 as the first LLM marketplace”
- openrouter.ai/docs/faq checked September 26, 2026
Backs: Free credits
“All new users receive a small free allowance to test out OpenRouter.”
- openrouter.ai/models checked September 26, 2026
Backs: Largest context window, Input price from, Models offered, Output price from
“1.05M context”
- openrouter.ai/pricing checked September 26, 2026
Backs: Bring your own key, Fee model, Free tier
“BYOK Limits: Standard $25,000 of list price inference / month with no fees, 5% fee after”
- openrouter.ai/support checked September 26, 2026
Backs: OpenAI-compatible API, Support channels
“We implement OpenAI-compatible `/completions` and `/chat/completions`, plus `/api/v1/models`.”
- openrouter.ai/terms checked September 26, 2026
Backs: Headquarters, Legal entity
“We operate the Service from our offices in New York”
- trust.openrouter.ai checked September 26, 2026
Backs: Compliance certifications
“SOC 2 Type 2”
Compiled from 7 source pages, checked September 26, 2026. We did not open an account.
OpenRouter
One key and one bill for teams routing across many models and providers.