> Hugging Face
Hugging Face Inference Providers routes one API key across 200+ models from multiple inference providers at the providers' own rates — no markup. Every user gets monthly credits ($0.10 on the Free plan; $2.00 PRO; $2.00/seat Team & Enterprise) applied automatically before pay-as-you-go; a BYOK mode routes through your own provider keys instead.
// pros
- One key across 200+ models from multiple providers — switching providers is a model-ID change, not a new account.
- Monthly credits for every user: $0.10 (Free), $2.00 (PRO), $2.00/seat (Team/Enterprise), applied automatically before any pay-as-you-go billing.
- Explicit no-markup billing: Hugging Face charges the providers' rates exactly — the aggregator adds convenience, not cost.
- BYOK mode: set a custom provider key and the same API passes through to your provider account, unbilled by HF.
- Org controls: centralized billing via X-HF-Bill-To, spend limits, per-provider toggles and usage breakdowns by model and provider.
// cons
- $0.10/month of free credits is a taste, not a tier — anything non-trivial moves to pay-as-you-go immediately.
- The Free-plan credit line is published as 'subject to change' — HF can alter or withdraw it.
- Custom-provider-key mode forfeits the monthly credits — the billing mode you pick determines whether the free allowance exists at all.
- The legacy hf-inference service is now mostly CPU inference — GPU workloads must route through the provider catalog.
Treat it as a routing layer with a coffee's worth of monthly free tokens: ideal for prototyping across many open models under one billing line, and for orgs that want consolidated inference spend with limits.
written by Mike Fleming, Uprouter editorial · verified against official sources September 16, 2026
Value taken from the provider’s own published free-tier terms, expressed in USD.
note: Official docs: $0.10/month Inference-Providers credits for Free users (PRO $2.00/month, Team/Enterprise $2.00/seat), credited monthly and applied before pay-as-you-go (confirmed 2026-09-16).
| Plan | Type | Monthly | Input /1M | Output /1M | Note |
|---|---|---|---|---|---|
| Pay-as-you-go | payg | — | — | — | provider rates + platform fee |
| Free tier | free | — | Free | Free | $0.10/mo |
| Subscription | subscription | — | — | — | $2/mo credits + 20x limits |
| Free tier (source-reported) | free | — | — | — | Monthly free credits: ~$0.10 for Free users (official notes subject to change), $2.00 for PRO, $2.00/seat for Team/Enterprise; Serverless Inference has free allowance for models under 10GB (some popular larger models also supported). |
| Free | free | — | — | — | HuggingChat is free with usage/message limits and periodic resets. |
| Inference Endpoint (T4) | payg | — | — | — | $0.50 per hour for NVIDIA T4 instances. |
| CPU Basic Space | free | — | — | — | 2 vCPU, 16 GB Memory; FREE on-demand. |
Pricing normalized from public sources — always verify with the provider.
// model prices
| Model | Input /1M | Output /1M | Context |
|---|---|---|---|
| Llama family | — | — | — |
| Qwen family | — | — | — |
| Gemma family | — | — | — |
| Mistral family | — | — | — |
| DeepSeek family | — | — | — |
| Meta: Llama 3.1 405B Instruct | — | — | 128k |
| Qwen: Qwen3 72B | — | — | 131k |
| DeepSeek: DeepSeek Coder V2 Instruct | — | — | 128k |
$ Does Hugging Face have a free tier?
Yes — Hugging Face offers a free tier that we value at roughly $0.10 in free credits. Always confirm current limits on Hugging Face's official pricing page — free tiers change without notice.
$ How much does Hugging Face cost per 1M tokens?
Hugging Face does not publish simple per-1M-token pricing (it may bill per request, per output, or via a subscription plan). Check their official pricing page for exact figures.
$ Is Hugging Face safe to use?
Uprouter rates Hugging Face as medium risk. Low risk: the aggregator states no-markup billing explicitly, publishes its credit table with the 'subject to change' caveat, and lets you exit HF billing entirely via BYOK. The real risks are Free-tier credit volatility and provider-side limits you don't directly control when HF routes the 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 | 15ms |
| 2d ago | up | 12ms |
| 3d ago | up | 17ms |
| 5d ago | up | 15ms |
| 6d ago | up | 15ms |
| 8d ago | up | 14ms |
| 9d ago | up | 17ms |
| 11d ago | up | 19ms |
| 12d ago | up | 33ms |
| 13d ago | up | 17ms |
| 14d ago | up | 16ms |
| 15d ago | up | 15ms |
| 16d ago | up | 15ms |
| 16d ago | up | 16ms |
| 17d ago | up | 12ms |
| 18d ago | up | 15ms |
| 19d ago | up | 15ms |
| 19d ago | up | 24ms |
| 20d ago | up | 14ms |
| 21d ago | up | 22ms |
| 22d ago | up | 11ms |
| 23d ago | up | 14ms |
| 23d ago | up | 20ms |
| 24d ago | up | 15ms |
| 25d ago | up | 32ms |
| 26d ago | up | 15ms |
| 27d ago | up | 18ms |
| 27d ago | up | 16ms |
| 28d ago | up | 17ms |
| 29d ago | up | 22ms |
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 risk: the aggregator states no-markup billing explicitly, publishes its credit table with the 'subject to change' caveat, and lets you exit HF billing entirely via BYOK. The real risks are Free-tier credit volatility and provider-side limits you don't directly control when HF routes the 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
# Hugging Face's own API — get a key from their dashboard
curl https://huggingface.co/chat/chat/completions \
-H "Authorization: Bearer THEIR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "llama-family",
"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.