models/NVIDIA/Cosmos Predict 2.5 2B Distilled
NVIDIA

Cosmos Predict 2.5 2B Distilled

video
Compare

Generate video from text and videos using NVIDIA's 2B Cosmos Distilled Model

MODALITIES
video

Cosmos Predict 2.5 2B Distilled (nvidia/cosmos-predict-2.5-distilled-text-to-video) is a video model from NVIDIA. Pricing via AIgateway: see pricing table. Call it via https://api.aigateway.sh/v1/videos/generations — set model="nvidia/cosmos-predict-2.5-distilled-text-to-video".

model · nvidia/cosmos-predict-2.5-distilled-text-to-video

Use this model

model: nvidia/cosmos-predict-2.5-distilled-text-to-video
curl https://api.aigateway.sh/v1/videos/generations \
  -H "Authorization: Bearer $AIGATEWAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"nvidia/cosmos-predict-2.5-distilled-text-to-video","prompt":"a slow drone shot over a mountain lake"}'
API schema

Call Cosmos Predict 2.5 2B Distilled from any OpenAI SDK

POST https://api.aigateway.sh/v1/videos/generations·Content-Type: application/json·Auth: Bearer sk-aig-...

Request body

json
{
  "model": "nvidia/cosmos-predict-2.5-distilled-text-to-video",
  "prompt": "A drone shot of a mountain lake at golden hour",
  "duration": 5,          // seconds
  "aspect_ratio": "16:9"
}

// Response is an async job — poll /v1/jobs/<id> until status === "completed".

Response

json
{
  "id": "job_abc123",
  "status": "queued",        // queued | processing | completed | failed
  "model": "nvidia/cosmos-predict-2.5-distilled-text-to-video",
  "created": 1776947082
}

// After completion:
{
  "id": "job_abc123",
  "status": "completed",
  "result": {
    "url": "https://media.aigateway.sh/video/abc123.mp4",
    "duration": 5,
    "resolution": "1920x1080"
  }
}

Quickstart

# See docs at https://aigateway.sh/docs

Errors

401authentication_errorInvalid or missing API key
402insufficient_creditsWallet empty (PAYG only)
404not_foundUnknown model or endpoint
429rate_limit_errorOver per-minute limit — see Retry-After header
500server_errorUpstream provider failed (retryable)
503service_unavailableUpstream saturated (retryable)
Full docs →API reference →OpenAPI spec →llms.txt →

Frequently asked questions

What is Cosmos Predict 2.5 2B Distilled?
Generate video from text and videos using NVIDIA's 2B Cosmos Distilled Model It is a video model from NVIDIA, accessible via AIgateway's OpenAI-compatible API at slug nvidia/cosmos-predict-2.5-distilled-text-to-video.
How much does Cosmos Predict 2.5 2B Distilled cost via AIgateway?
Pass-through pricing. See the pricing panel on this page for exact rates.
How do I call Cosmos Predict 2.5 2B Distilled from my code?
Point the OpenAI SDK at https://api.aigateway.sh/v1 with your AIgateway key and set model to "nvidia/cosmos-predict-2.5-distilled-text-to-video". The request and response shapes match OpenAI exactly.
Does Cosmos Predict 2.5 2B Distilled support streaming, tool calling, vision, and JSON mode?
Streaming — no. Tool calling — no. Vision — no. JSON mode — no. Prompt caching — no.
Can I bring my own NVIDIA API key (BYOK)?
Yes. Attach a NVIDIA key in your AIgateway dashboard and this model flips to pass-through — you pay NVIDIA directly and AIgateway adds no platform fee on those calls.