Speed
Throughput
Nine GPUs. One answer.
True becauseThe 9-GPU pool measured ~1,195 tokens/s; ~85 tok/s on a single stream.
Private, US-hosted GPUs behind an OpenAI-compatible endpoint. Point your existing tooling at one new base URL and run models your laptop can’t — with bigger models on higher plans — at a flat monthly price, with no per-token surprise bills and zero data retention. We never train on your prompts.
# Your existing OpenAI code — change ONE line:
from openai import OpenAI
client = OpenAI(
base_url="https://api.nexusaisolutions.io/v1", # ← only this
api_key="nxs_your_key",
)
r = client.chat.completions.create(
model="qwen2.5-32b",
messages=[{"role":"user",
"content":"Runs on hardware you can audit."}],
)
# Same SDK. Same shapes. Your data never leaves the cluster.A laptop runs small models and throttles under sustained load. Nexus gives you real capacity on private GPUs we own, at a flat monthly price, without your prompts becoming someone's training data.
Shared models on call up to 32B-class on Solo and Studio, up to 70B-class on Team, plus your own private serve up to ~14B on Solo. The fleet has benchmarked 123B; larger models are served on request. Llama, Qwen, Mistral, DeepSeek, or your own fine-tuned adapter.
Measured ~85 tok/s on a single stream (7B), prefill above 1,200 tok/s, and ~1,195 tok/s across the 9-GPU serving pool. On a private network there's no public-internet round trip; latency is just model speed.
No prompts, no completions, no telemetry leave the cluster. We don't train on your data and we don't retain it. Your fine-tuned adapter stays on your tenant — never pooled, never published.
Each picture maps to something the system does today, so it never promises more than the product.
Nine GPUs. One answer.
True becauseThe 9-GPU pool measured ~1,195 tokens/s; ~85 tok/s on a single stream.
Set a cap. Never get surprised.
True becauseChat, image, voice, video and 3D all draw from one pool of GPU-minutes, with a cap you set.
Bring your own. Keep your own.
True becauseCatalogs mark Shared (included) vs Yours (uploaded or trained), and yours stay in your tenant.
If your tool speaks OpenAI — Cursor, Continue, Open WebUI, LangChain, your own harness — it already speaks Nexus. Change the base URL, keep everything else.
Full model list & endpoints in the docs.
Drag your monthly usage and pick the model you’d otherwise pay for. We show your flat Nexus cost, your effective per-token rate here, the metered cost of the same volume, and exactly where the line crosses — honestly, both ways.
≈ light chat 1-2M · heavy coding assistant 10-30M · batch/agents 50M+
Published list prices, Sept 2026 — verify current. Edit to your real rate.
3,000 GPU-min included · ~23M tokens
Anchor: ~8,000 output tokens per GPU-minute (blended 7B, measured). 13B/30B consume more GPU-min per token; the meter self-adjusts. Overage is capped by you: set a cap, or auto-upgrade.
What metered APIs can’t give you at any price:
Pick a tier at signup. Solo trials free for 7 days — no card. Studio trials 7 days with a card on file ($0 today, charged on day 7 unless you cancel). Team is paid from day one, with no trial. Already decided? Start paid and skip the trial. No free-forever tier.
Indie developers & solo builders
Small teams & one-person agencies
Growing product or agency
A laptop runs small models and throttles under sustained load. Nexus gives you private GPUs, bigger models as your plan grows, and real concurrency, without buying and cooling your own card.
No. No training, no retention, no telemetry. Prompts and completions stay on your tenant and are wiped at job end. A fine-tune you run stays a private adapter on your tenant.
Llama 3.x, Qwen 2.5, Mistral, DeepSeek, and your own fine-tuned variants. Solo reaches shared models on call up to 32B-class (reasoning, coding and vision models start when you need them) plus one private serve up to ~14B. Studio gets the same shared models plus up to 2 dedicated serves, and downloads from Civitai, Tensor.Art and Hugging Face. Team goes up to 70B-class shared with multi-GPU serves (30B-class dedicated). Enterprise gets larger models served on request.
Your plan includes a monthly GPU-minute bucket, and overage is capped by you: set a cap so the bill never runs past it, or auto-upgrade to the next tier. No opaque per-token metering, no throttling mid-project.
Sign up for Solo (no card), get an nxs_… key, point your base_url at our endpoint. Minutes, not a sales cycle.
Solo is 7 days free, no card. Bring your OpenAI code and change one line.
Try Solo free — no card requiredUS-hosted private GPUs · your data never leaves the cluster