{"rid":"inference/hetzner-inference-api","card":{"rid":"inference/hetzner-inference-api","eid":"hetzner","topic":"inference","slug":"hetzner-inference-api","provider":"Hetzner Inference API (experiment)","url":"https://docs.hetzner.com/general/company-and-policy/experiments/inference/","domain":"hetzner.com","category":"inference endpoint, experimental free API","tier":null,"verdict":"qualifies-with-caveat","vorder":1,"grp":"usable","confidence":"reported","binding":"request-rate","identity_kind":"card","identity_basis":"defaulted","needs_card":1,"needs_phone":null,"needs_id":null,"needs_institution":null,"country":null,"crypto":null,"interfaces":"rest","egress_metered":null,"per_object_bytes":null,"retention_days":null,"verified_at":"2026-10-28T04:36:18.593Z","verified_basis":"live","checked_on":"2026-10-29","stale_on":"2026-11-13","n_evidence":6,"usd_month":null,"usd_per_tb_month":null,"summary":"free of charge while the Inference API remains in experimental status; two models, Qwen/Qwen3.6-35B-A3B-FP8 and Qwen3.8-27B, each with a 262,144-token contex...","duplicate_of":null,"n_unknown":5,"free_unit":"tokens"},"record":{"topic":"inference","slug":"hetzner-inference-api","provider":"Hetzner Inference API (experiment)","url":"https://docs.hetzner.com/general/company-and-policy/experiments/inference/","category":"inference endpoint, experimental free API","free_tier":{"stated":"free of charge while the Inference API remains in experimental status; two models, Qwen/Qwen3.6-35B-A3B-FP8 and Qwen3.8-27B, each with a 262,144-token context and text plus image input; limits per API key are 4M input tokens and 100k output tokens per 60 s and 10 requests per 60 s","unit":"tokens","metered_on":"request-count","source_url":"https://docs.hetzner.com/general/company-and-policy/experiments/inference/","read_on":"2026-10-28","notes":"the vendor says it will email users before the experimental status changes; the model list changes during the experiment and GET /v1/models is definitive"},"per_object_cap":{"stated":"not applicable"},"retention":{"stated":"not applicable"},"egress":{"terms":null,"metered":false},"interface":["rest"],"tooling":"any OpenAI-compatible client at https://inference.hetzner.com/api/v1; endpoints /v1/models, /v1/completions and /v1/chat/completions","chaining":"Hetzner-hosted open-weight models behind a per-key ceiling of 10 requests per minute","limits":[{"kind":"request-rate","value":"per API key: 10 requests per 60 s, 4M input tokens per 60 s, 100k output tokens per 60 s; HTTP 429 beyond any of them","source":"https://docs.hetzner.com/general/company-and-policy/experiments/inference/"}],"identity_cost":{"kind":"card","detail":"unverified at the vendor: the docs page names no sign-up requirement. A third-party tracker reports that minting a token needs a Hetzner account whose payment verification can include a card charge or other routes","country":null},"reliability_note":"experimental service offered as is: performance and availability are not guaranteed, no backups, and the model list can change without a fixed schedule","confidence":"reported","confidence_note":"the rate limits and the free status are quoted from the vendor's docs page; the identity requirement is a third-party report because the key-minting page is a client-rendered application that could not be captured as text.","evidence":[{"url":"https://docs.hetzner.com/general/company-and-policy/experiments/inference/","date":"2026-10-28","proved":"the API is free while it stays experimental","capture":"sha256:4ee066eef1b4dbb2a1c5ade07bb0f6588e5c65952249cc6cd2df5b7668b63a9f","quote":"As long as the Inference API remains in experimental status, it is free of charge."},{"url":"https://docs.hetzner.com/general/company-and-policy/experiments/inference/","date":"2026-10-28","proved":"per API key, 4M input tokens and 100k output tokens per 60 seconds","capture":"sha256:4ee066eef1b4dbb2a1c5ade07bb0f6588e5c65952249cc6cd2df5b7668b63a9f","quote":"| 60s | 4M | 100k |"},{"url":"https://docs.hetzner.com/general/company-and-policy/experiments/inference/","date":"2026-10-28","proved":"request-level limit of 10 requests per 60 seconds","capture":"sha256:4ee066eef1b4dbb2a1c5ade07bb0f6588e5c65952249cc6cd2df5b7668b63a9f","quote":"| 60s | 10 |"},{"url":"https://docs.hetzner.com/general/company-and-policy/experiments/inference/","date":"2026-10-28","proved":"Qwen3.8-27B is served with a 262,144-token context and image input","capture":"sha256:4ee066eef1b4dbb2a1c5ade07bb0f6588e5c65952249cc6cd2df5b7668b63a9f","quote":"| Qwen3.8-27B | Dense | 262,144 tokens | Text, Image |"},{"url":"https://docs.hetzner.com/general/company-and-policy/experiments/inference/","date":"2026-10-28","proved":"no availability guarantee for the experimental service","capture":"sha256:4ee066eef1b4dbb2a1c5ade07bb0f6588e5c65952249cc6cd2df5b7668b63a9f","quote":"Performance and availability are not guaranteed, especially during periods of high demand."},{"url":"https://mvalentsev.github.io/awesome-free-ai-coding/providers/hetzner-inference/","date":"2026-10-28","proved":"third-party report on what getting a key requires","capture":"sha256:a696d1e534ee123752b17c708bcaa98d84a433899e18455b82e7c5f11337febb","quote":"Minting a token needs a Hetzner account; payment verification can include a card charge or other routes."}],"verdict":"qualifies-with-caveat","verified_at":"2026-10-28T04:36:18.593Z","binding":"request-rate","binding_note":"10 requests per 60 s per key is the ceiling that decides throughput; the token windows are far above it","notes":"the identity requirement is the open item: the kind is recorded as card because the only source says payment verification can include a card charge.","endpoint":"https://inference.hetzner.com/api/v1/chat/completions"},"claims":[{"path":"$.free_tier.stated","kind":"free-quota","num":10,"unit":"requests","per":null,"currency":null,"usd":null,"fx_date":null,"parse":"ok"},{"path":"$.limits[0].value","kind":"limit:request-rate","num":10,"unit":"requests","per":null,"currency":null,"usd":null,"fx_date":null,"parse":"ok"}],"unknown":[["$.egress.metered","egress-unknown"],["$.egress.terms","process"],["$.identity_cost.kind","identity-defaulted"],["$.per_object_cap.stated","not-applicable"],["$.retention.stated","not-applicable"]],"history":[],"drift":[]}