One key. Every media model.

Segmind AI Gateway
for Media Models.

One API for image, video, and audio across every major lab. Automatic failover, BYOK, list-price billing. No markup.

segmind-gateway
Modality
curl -X POST "https://api.segmind.com/v1/nano-banana-pro" \
  -H "x-api-key: $SEGMIND_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "an editorial portrait, golden hour"
  }'
One endpoint. Swap nano-banana-pro for any of 500+ media models.
Routing500+ models · 40+ providers
Seedream 5.0
Veo 3.1
Kling 3.0
Flux Pro
GPT Image 1.5
ElevenLabs v3
Runway Gen-4
Seedance 2.0

Segmind

Gateway

Live

Segmind AI Gateway is an AI gateway for media models: one REST API and one API key for 500+ image, video, audio, voice and language models from 40+ AI labs, including Google, OpenAI, ByteDance, Alibaba and ElevenLabs. Fast models answer in a single call, long jobs such as video run on the async API, and billing is pay as you go with no charge for failed requests.

Why a Gateway

Three things you stop building
the day you switch to Segmind.

One key, every media model

Stop wrangling 40+ providers.

One API key, one billing relationship, one set of credentials. Image, video, audio, voice. Every major lab behind a single endpoint with consistent request and response shapes.

Automatic failover

When a provider goes dark, your app doesn’t.

Configure a fallback chain across providers. If a primary model is rate-limited, regionally degraded, or returning errors, the gateway retries on the next model, without you touching code.

No platform fees

Pay exactly what providers charge.

No gateway tax, no per-request surcharge, no inflated token rates. Per-generation cost is itemized in the dashboard, down to the second of video.

Media-Native Infrastructure

The things LLM gateways
weren’t built to do.

Async jobs, built in

Long generations are first-class.

Five-minute video generations don’t fit the LLM token-streaming model. The gateway treats every job as durable: poll, webhook, or stream progress updates. No request timeouts.

Job · gw_8f2a1c0erunning · 4m12s
queued
0.4s
generating · veo-3.1
3m 48s
uploading · S3
-
S3-backed delivery

Outputs land in object storage.

Generated images and videos are uploaded to durable storage with signed URLs. No 30 MB base64 blobs in your response payloads.

output_url:
https://cdn.segmind.com/generations/gw_8f2a1c0e/video.mp4
Webhooks

Fire-and-forget jobs.

Pass a callback URL. Signed delivery on completion, automatic retry with backoff.

POSTyour-app.com/hooks200 OK
{
  "event": "job.succeeded",
  "id": "gw_8f2a1c0e",
  "output_url": ".../video.mp4"
}
Per-generation cost

Every generation, priced.

Itemized down to the second of video. See what each model actually costs you.

5s · 1080plast 24h
Seedance 2.0
$0.36
Veo 3.1
$1.20
Kling 3.0 Pro
$0.54
BYOK at 0%

Bring your own keys.

Plug in existing provider keys. We route, observe, and bill at zero markup.

Connected providers0% markup
OOpenAI

sk-•••a7e2

GGoogle

AIza•••82f

RRunway

run-•••3d1

EElevenLabs

el-•••9ce

Rate-limit smoothing

No more provider 429s.

The gateway queues, throttles, and re-routes around provider rate limits. Your app sees a steady-state API, not provider edge cases.

Steady 60 RPM, smoothed across 3 providers

Observability

Every generation, fully traced.

Input, provider, output, with cost, latency, and model version pinned. What LLM gateways pioneered, extended for media: prompt, seed, output URL, p95.

Trace · gw_8f2a1c0e8.4s · $1.20
input"cinematic shot, low golden light, 35mm…"
providerveo-3.1
outputvideo.mp4

How Routing Works

Never bet on one model.

Provider outages are someone else’s problem.

STEP 01

Primary

Your call hits the model you specified. In the happy path, that’s where it ends.

STEP 02

Fallback chain

If the primary is rate-limited, regionally degraded, or returning errors, we retry across your configured fallback chain (e.g., Veo → Kling → Runway).

STEP 03

BYOK or credits

Each hop uses your own provider key if you’ve plugged one in. Otherwise we use Segmind credits at the same list price.

Example config

{
  "model": "veo-3.1",
  "fallback": ["kling-3.0-pro", "runway-gen-4"],
  "byok": { "google": "gky_…", "runway": "rwk_…" }
}

Observability

See every generation.
Cost, latency, output, in one place.

Gateway Dashboard· last 30 days
Live
Generations

127.5K

+12.5% vs last month

Avg Latency

14.2s

all models

Fastest Model

0.8s

embedding · avg

By model

ModelGenerationsShareAvg Latency
nano-banana48,72038.2%11.8s
segmind-vega28,69622.5%2.2s
face-to-many18,87614.8%9.1s
faceswap-v516,07012.6%8.3s
seedance-1.5-pro15,17811.9%77.9s

Sample data. Filter by model, status, source and date range. Export to CSV.

Positioning

Built for media.
Not just text.

Generic LLM gateways are great for chat. The shape of a media workload, multi-minute generations, multi-gigabyte outputs, per-second billing, is a different problem.

CapabilitySegmind (media)Generic LLM gateway
Long-running generations (1 to 5 min)KEY
Async job, durable
Times out
Output delivery
Signed S3 URLs
Inline tokens
Per-second video pricingKEY
Native, itemized
Token-based only
Webhook callbacks
Job-level events
Streaming chunks
Image / video / audio in one APIKEY
Yes, native
Text-first
Token streaming for chat
Not applicable
First-class
Prompt caching for chat
Not applicable
First-class
Provider failover
Across labs and models
Across models
BYOK at 0% markup
Yes
Varies

If your workload is chat, an LLM gateway is the right choice. If you’re shipping image, video, or audio at scale, you’ll save a lot of glue code by starting here.

Who builds on Segmind

One gateway.
Every team shipping AI media.

From two-person startups to platforms serving millions. Same API, different scale.

STARTUPS

Ship the feature, not the plumbing.

One API, every frontier model across image, video, and audio. Skip the integration tax, the cold starts, the per-lab contracts. Go from idea to production in an afternoon, and let the gateway handle scale when you get there.

See the developer story

Models

500+

Time to first call

< 5 min

Billing

Pay as you go

Key facts

Models
500+ image, video, audio, voice, 3D and language models (as of 13 Oct 2026)
Model makers
40+ AI labs, including Google, OpenAI, ByteDance, Alibaba, Kling and ElevenLabs (as of 13 Oct 2026)
Endpoint
POST https://api.segmind.com/v1/<model-slug> (sync) or /v2/<model-slug> (async)
Authentication
One Segmind API key, sent in the x-api-key header
Long jobs
Async v2 API: submit, get a request_id at once, then poll the status URL
SDK
Official Python SDK (pip install segmind); any other language over plain HTTPS
Billing
Pay as you go, per generation or per GPU-second; credits from $10
Failed requests
Never charged, whatever the cause
Rate limits
60 to 1,000 requests per minute, depending on plan (as of 13 Oct 2026)
Output storage
Generated files not deleted automatically; request history keeps requests 7 days; uploaded inputs deleted after 2 days
Monitoring
Per-request log, cost by model and API key, and API health dashboards in the console
Enterprise plan
Custom uptime SLA, dedicated Slack support

Pricing

Pay exactly what providers charge. No platform fees.

Provider list price
No platform fees
No request fees
Pay as you go

Frequently asked questions

What is the Segmind AI Gateway?

The Segmind AI Gateway is one REST API for 500+ image, video, audio, voice and language models from 40+ AI labs. You call every model with the same Segmind API key at https://api.segmind.com/v1/<model-slug>, switch models by changing the slug, and pay for each generation from one credit balance.

Which languages and SDKs work with the Segmind AI Gateway?

Any language that can send an HTTPS request works with the Segmind AI Gateway, including cURL, Python, Node.js, Go and Rust. Segmind also publishes official SDKs: Python (pip install segmind) handles submitting, polling and file uploads, and JavaScript (npm install segmind, Node.js 22+) offers segmind.run(), which queues a request and waits for the result. JavaScript docs: https://docs.segmind.com/docs/sdks/javascript

How does the Segmind AI Gateway handle long video generations?

Long jobs run on the Segmind AI Gateway's async v2 API. You POST to https://api.segmind.com/v2/<model-slug>, get a request_id straight away, and poll the status URL until the result is ready, so no HTTP connection stays open. The sync v1 API suits fast models that finish in a few seconds.

How much does the Segmind AI Gateway cost?

The Segmind AI Gateway is pay as you go: buy credits from $10, and each model is billed at its own price, per generation or per GPU-second. Paid plans add monthly credits and higher rate limits. Failed requests are never charged, and each successful response reports its exact cost in the x-cost header or the metrics.cost field.

What are the Segmind AI Gateway rate limits?

Segmind AI Gateway rate limits depend on the plan: 60 requests per minute on Flexible, 120 on Pro, 500 on Business and 1,000 pooled on Scale, as of October 2026. A request over the limit returns HTTP 429, so wait briefly and retry. Contact Segmind sales for a higher limit.

How long does the Segmind AI Gateway keep generated files?

The Segmind AI Gateway does not delete generated files automatically: they stay stored after a request leaves request history, which lists each request for 7 days. Save the output URL or download the file when you fetch the result. Input files you upload are kept for 2 days by default, then deleted.

Can I see the cost and latency of each AI Gateway request?

Yes. The Segmind console lists every AI Gateway request in Generations with its model, status, duration, cost and request ID, filterable by model, status, source and date, with CSV export. Cost Analytics breaks spend down by model, source, API key and team member, and API Health shows success rate and latency percentiles.

Does the Segmind AI Gateway come with an SLA?

The Segmind AI Gateway's Enterprise plan comes with an uptime SLA agreed with sales, together with dedicated Slack support. The Flexible, Pro, Business and Scale plans do not list an SLA; their support runs from the community Discord to a 1 business day response on Scale.

Available now on Segmind

One API key.
Every media model.

Start building in under five minutes. No contract, list-price billing, pay as you go.

Segmind is an AI infrastructure platform with 500+ media models from 40+ providers.
Built for image, video, audio, and voice workloads at production scale.

Last updated