> NVIDIA NIM
NVIDIA NIM's hosted API (build.nvidia.com / integrate.api.nvidia.com) serves a broad catalog of open-weight models — Nemotron, Llama, Qwen, DeepSeek, GLM and more — in an OpenAI-compatible shape, with a permanent free tier: API key at sign-up, no card, 40 requests/minute by default (200 RPM granted on request, per NVIDIA's developer forums).
// pros
- A genuinely permanent free tier: API key at sign-up, no card, no published trial countdown — 40 requests/minute by default.
- The 40 RPM default is raisable: NVIDIA's own developer forum shows support granting 200 RPM on request.
- Broad open-model catalog (12+ models tracked here) — Nemotron, Llama, Qwen, DeepSeek, GLM and more behind one OpenAI-compatible key.
- RPM (not token) metering favors high-frequency, small requests — well suited to agent loops and tool-calling traffic.
- First-party from the model vendors' silicon partner, so catalog models match the hardware they're tuned for.
// cons
- NVIDIA publishes no single canonical free-tier limits page (docs URLs 404 as of 2026-09-16) — the 40 RPM default is documented through developer-forum support responses, so plan for the policy to move.
- Token-level allowances beyond the RPM cap aren't clearly published — instrument your own usage to know the true ceiling.
- The catalog is a rolling partner set — models appear and retire without deprecation notices as formal as a lab's own.
- The hosted free tier is dev-oriented: no SLA, and production guidance points to NIM microservices on your own infrastructure or paid channels.
A solid zero-commitment baseline: take a key, use the 40 RPM, and request 200 RPM if the workload justifies it. For production, treat the hosted API as the dev channel and plan NIM on your own infra or a partner route.
written by Mike Fleming, Uprouter editorial · verified against official sources September 16, 2026
Free tier present — value not quantified
This provider offers a free tier, but its terms don’t convert cleanly into a USD figure (e.g. request-count limits or capacity-dependent pools). We refuse to print a misleading zero here. See the note below and the official pricing page for the exact limits.
note: Free tier is rate-limited (40 RPM default, 200 on request) rather than credit-based — no USD value to quantify.
| Plan | Type | Monthly | Input /1M | Output /1M | Note |
|---|---|---|---|---|---|
| Free tier (source-reported) | free | — | — | — | Permanent free tier — signup yields an API key, no credit card, no trial countdown; current limit ~40 RPM (can request 200 RPM). Historically granted 1000 credits (enterprise email +4000); some users hit credits zeroing out unused and 402 'Cloud credits expired' errors. |
| Free Tier | free | — | — | — | 40 requests per minute permanent free limit for hosted models. |
Pricing normalized from public sources — always verify with the provider.
// model prices
| Model | Input /1M | Output /1M | Context |
|---|---|---|---|
| Nemotron 3 Nano 30B A3B | $0.050 | $0.200 | 262k |
| Nemotron 3.5 Lightning | $0.080 | $0.200 | 262k |
| Nemotron 3 Super | $0.085 | $0.400 | 262k |
| Nemotron 3 Ultra | $0.625 | $3.13 | 256k |
| nvidia/llama-3.1-nemotron-ultra-253b-v1 | — | — | — |
| mistralai/mistral-large-3-675b-instruct-2512 | — | — | — |
| qwen/qwen3.5-397b-a17b | — | — | 262k |
| deepseek-v4-pro | — | — | 1.0M |
| GLM 5.2 | — | — | 1.0M |
| gpt-oss-120b | — | — | 131k |
| kimi-k2.6 | — | — | 262k |
| Llama 3.3 70B Instruct | — | — | 131k |
$ Does NVIDIA NIM have a free tier?
Yes — NVIDIA NIM offers a free tier. Always confirm current limits on NVIDIA NIM's official pricing page — free tiers change without notice.
$ How much does NVIDIA NIM cost per 1M tokens?
The cheapest model we track at NVIDIA NIM is $0.050 per 1M input tokens and $0.200 per 1M output tokens. This is normalized from NVIDIA NIM's published pricing — verify with the provider before purchasing, since prices change frequently.
$ Is NVIDIA NIM safe to use?
Uprouter rates NVIDIA NIM as low risk. Low data-integrity risk — it's NVIDIA's first-party API — but the documentation is the caveat: the free tier's exact limits live in forum support responses rather than a canonical pricing page. Re-verify the RPM before building around it (verified 2026-09-16: 40 RPM default, 200 RPM on request). Risk ratings are editorial, evidence-based, and never influenced by affiliate relationships.
Raw community sentiment, shown unblended. Votes never feed into objective data fields like price, risk, or status — and never into sort order.
No published reviews yet — be the first.
// more_informationprice history · live status · risk & confidence · code snippets
price_history
No recorded changes yet.
live_status
full history →best-effort probes · not a guarantee · last 15h ago
| Checked | Status | Latency |
|---|---|---|
| 15h ago | up | 13ms |
| 2d ago | up | 18ms |
| 3d ago | up | 14ms |
| 4d ago | up | 19ms |
| 6d ago | up | 14ms |
| 7d ago | up | 13ms |
| 9d ago | up | 16ms |
| 11d ago | up | 13ms |
| 12d ago | up | 16ms |
| 13d ago | up | 26ms |
| 14d ago | up | 17ms |
| 15d ago | up | 12ms |
| 15d ago | up | 18ms |
| 16d ago | up | 14ms |
| 17d ago | up | 12ms |
| 18d ago | up | 11ms |
| 19d ago | up | 15ms |
| 19d ago | up | 14ms |
| 20d ago | up | 14ms |
| 21d ago | up | 15ms |
| 22d ago | up | 20ms |
| 22d ago | up | 13ms |
| 23d ago | up | 17ms |
| 24d ago | up | 14ms |
| 25d ago | up | 13ms |
| 26d ago | up | 14ms |
| 27d ago | up | 14ms |
| 27d ago | up | 15ms |
| 28d ago | up | 14ms |
| 29d ago | up | 13ms |
Uprouter runs its own probes; best-effort, not a guarantee. Probe results depend on our network location and cadence — for production, always use the provider’s own status page.
risk_&_confidence
Low data-integrity risk — it's NVIDIA's first-party API — but the documentation is the caveat: the free tier's exact limits live in forum support responses rather than a canonical pricing page. Re-verify the RPM before building around it (verified 2026-09-16: 40 RPM default, 200 RPM on request).
Confidence reflects how much verifiable evidence (official docs, probe history, community reports) backs this entry. Risk is our editorial assessment of operator reliability and terms — not financial advice.
code_snippets
# NVIDIA NIM's own API — get a key from their dashboard
curl https://YOUR-PROVIDER-ENDPOINT/chat/completions \
-H "Authorization: Bearer THEIR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "nvidia-llama-3-1-nemotron-ultra-253b-v1",
"messages": [{ "role": "user", "content": "Hello" }]
}'Direct to the provider — Uprouter receives nothing; verify their current pricing before purchasing.
Ranked by neutral similarity signals only — live status, free-tier shape and free-credit proximity. Commercial relationships never influence this list.