Skip to content
01

Providers + pricing

CapabilitiesProvider details
● Credits · billed by AnyRouter
NVIDIA
nvidia
$0
$0
● Your own key · billed by your provider · AnyRouter fee $0
NVIDIA
nvidia-byok
$0
$0
Unavailable
OpenRouter
openrouter-byok
$0
$0
Unavailable
● Free pool · donated keys · not available for this model — donate a key
02

Try it

POST /api/v1/chat/completions
import { createAnyRouter } from "@anyr/ai-sdk-provider"import { generateText } from "ai" const anyrouter = createAnyRouter() const { text } = await generateText({  model: anyrouter("nvidia/cosmos3-nano-reasoner"),  prompt: "Say hi in 3 words.",})console.log(text)
Set ANYROUTER_API_KEY · edits on the demo update this code
03

Uptime + latency

—recent checks · all routes

No health checks recorded for this model yet.

24h
No traffic in the last 24h.
More performance detail
Usage analytics

Loading usage…

Uptime & Health
No uptime data yet

These providers haven't been health-probed for this model yet. The router still routes around upstreams that fail live requests — uptime fills in once probe history accrues.

Cosmos 3 Nano ReasonerDisabled

NVIDIA has not exposed this on the hosted API yet (integrate.api.nvidia.com 400s); it ships as self-host NIM + HF weights only.

  • NVIDIA
  • NVIDIA
  • OpenRouter
Visiontext + image + video → text128Ktoken context
Context
128K
Input
$0
Output
$0
TTFT
—
Uptime
—
Routes
3
04

Your access

Credits
Your own key · AnyRouter fee $0

Run Cosmos 3 Nano Reasoner on your own key — your requests are billed by the provider. Pool callers pay AnyRouter credits.

No BYOK keys configured for this model yet.

Share a key with the pool to earn credits for every request it serves, covering your plan cost.

Free poolnot in pool yetDonate key →
Create API key for this model
Text generation
Context length128,000 tokens
Max output102,400 tokens
ArchitectureTransformer
Categorymultimodal
ReleasedJun 4, 2026
Modalities
→
Capabilities
05

About

A vision language model that excels in understanding the physical world using structured reasoning on videos or images. Built on NVIDIA's Cosmos foundation for physical AI, it provides dense, step-by-step reasoning about visual scenes and temporal sequences. Served via NVIDIA NIM.

Released 2026-06-04 · params: max_tokens · temperature · top_p · stop · tools · tool_choice · response_format

API & code
Share cards
Cosmos 3 Nano Reasoner share card
Cosmos 3 Nano Reasoner
NVIDIA upstream share card
NVIDIA upstream
NVIDIA upstream share card
NVIDIA upstream
OpenRouter upstream share card
OpenRouter upstream
Back to models