In the EU tier, s1-pro uses mammoth-1 27B by baselabs (MIT); s1-fast and s1-vision use Blink v0.3 26B-A4B (Apache-2.0). In the EU tier, s1-pro is answered by mammoth-1 (s1-pro-5), or by s1-pro-3 (vLLM-SR d3) if mammoth-1 cannot answer. In the Peer-to-Peer tier, it is answered by s1-pro-3, with or without network processing enabled. The pinned id s1-pro-5 is EU tier only. model_version shows which version answered. Perplexity Decider v1.1 stays pinned as s1-pro-2 through 30 November 2026; s1-pro-2 and s1-pro-4 retire on 1 December 2026; s1-pro-3 stays available through at least 27 January 2027. Prices are per 1 million input tokens; output tokens are free. Free allowance by invitation. Pay as you go; top up from EUR 1. Private individuals and businesses worldwide can sign up and top up. Taxes are shown in checkout before you pay.
New EUR accounts: your first API key comes with a one-time EUR 0.50 credit (verified e-mail, limited to 36 credits per 7 days, valid 30 days, not combined with other credits; Terms 5.3(e)). At today's list price that is several million input tokens on s1-pro.
Current rates
These rates exclude applicable taxes. Top-ups are prepaid credit via Stripe with proper invoices from productivity-boost.com Betriebs UG (haftungsbeschränkt) & Co. KG. Private individuals and businesses worldwide can sign up and top up. Taxes are shown in checkout before you pay. Optional automatic top-up: in the dashboard you can let us charge your saved card when your balance falls below a threshold you set (amount from EUR 10, monthly cap you set, off by default, switch off any time); each charge gets a normal invoice with VAT. EUR rates are the approved list rates; USD rates are EUR × 1.1403 (ECB reference of 25 September 2026), rounded to $0.001 per million tokens.
s1-pro is our offer for speed, hosting in the EU, dedicated capacity and one API from one provider. The rates in the table apply to new customers. Existing customers keep $0.038 (Peer-to-Peer) and $0.040 (EU) per 1M input tokens until 12 November 2026, which is €0.033 and €0.035; existing customers are notified by email. Usage created earlier keeps the rate on its receipt.
USD rates
Model
Peer-to-Peer Elastic capacity · per-key opt-in
EU Hosting in the EU · DPA
s1-fast based on Blink v0.3 26B-A4B (Apache-2.0), NVFP4 · text
$0.019
$0.021
s1-pro mammoth-1 27B by baselabs (MIT), FP8 on our engine · text · In the EU tier, s1-pro is answered by mammoth-1 (s1-pro-5), or by s1-pro-3 (vLLM-SR d3) if mammoth-1 cannot answer. In the Peer-to-Peer tier, it is answered by s1-pro-3, with or without network processing enabled. The pinned id s1-pro-5 is EU tier only. model_version shows which version answered.
$0.080 Existing customers until 12 November 2026: $0.038
$0.084 Existing customers until 12 November 2026: $0.040
s1-pro-2 Perplexity Decider v1.1 27B (Apache-2.0), NVFP4 · previous s1-pro, pinned through 30 November 2026, retires 1 December 2026 · billed at the s1-pro list rate
$0.080
$0.084
s1-pro-3 vLLM-SR d3 26B (Apache-2.0), FP8 · pinned id · available through at least 27 January 2027
$0.080
$0.084
s1-pro-5 mammoth-1 27B by baselabs (MIT), FP8 on our engine · text · pinned id · EU tier only · billed at the s1-pro list rate
—
$0.084
s1-vision · text only same Blink model · text without an image
$0.031
$0.032
s1-vision · with image same Blink model · request contains one image; bills all input tokens (text and image positions)
$0.217
$0.228
reax-1 REAX 1 by Decision Models, a fine-tune of Perplexity Decider v1.2 by Perplexity (Apache-2.0), derived from Qwen3.8-27B by the Qwen team (Apache-2.0). Not affiliated with or endorsed by Perplexity AI or the Qwen team. · text · alias of the current REAX 1 release · billed at the s1-pro list rate
$0.080
$0.084
reax-1-20261021 pinned REAX 1 release of 21 Oct 2026 · billed at the s1-pro list rate
$0.080
$0.084
reax-1.3 REAX 1.3 by Decision Models · 1 question per request · 28,672 billed input tokens · EU only · output free · Model page
not offered
$0.400
reax-1 always points to the current REAX 1 release. We may replace the model behind this id with a newer REAX version at any time, with the same API and price; we switch only to a version that measures at least as well on our evaluations. For reproducibility, pin a dated id such as reax-1-20261021; pinned ids stay available for at least 30 days after they stop being current.
EUR rates
Model
Peer-to-Peer Elastic capacity · per-key opt-in
EU Hosting in the EU · DPA
s1-fast based on Blink v0.3 26B-A4B (Apache-2.0), NVFP4 · text
€0.017
€0.018
s1-pro mammoth-1 27B by baselabs (MIT), FP8 on our engine · text · In the EU tier, s1-pro is answered by mammoth-1 (s1-pro-5), or by s1-pro-3 (vLLM-SR d3) if mammoth-1 cannot answer. In the Peer-to-Peer tier, it is answered by s1-pro-3, with or without network processing enabled. The pinned id s1-pro-5 is EU tier only. model_version shows which version answered.
€0.070 Existing customers until 12 November 2026: €0.033
€0.074 Existing customers until 12 November 2026: €0.035
s1-pro-2 Perplexity Decider v1.1 27B (Apache-2.0), NVFP4 · previous s1-pro, pinned through 30 November 2026, retires 1 December 2026 · billed at the s1-pro list rate
€0.070
€0.074
s1-pro-3 vLLM-SR d3 26B (Apache-2.0), FP8 · pinned id · available through at least 27 January 2027
€0.070
€0.074
s1-pro-5 mammoth-1 27B by baselabs (MIT), FP8 on our engine · text · pinned id · EU tier only · billed at the s1-pro list rate
—
€0.074
s1-vision · text only same Blink model · text without an image
€0.027
€0.028
s1-vision · with image same Blink model · request contains one image; bills all input tokens (text and image positions)
€0.190
€0.200
reax-1 REAX 1 by Decision Models, a fine-tune of Perplexity Decider v1.2 by Perplexity (Apache-2.0), derived from Qwen3.8-27B by the Qwen team (Apache-2.0). Not affiliated with or endorsed by Perplexity AI or the Qwen team. · text · alias of the current REAX 1 release · billed at the s1-pro list rate
€0.070
€0.074
reax-1-20261021 pinned REAX 1 release of 21 Oct 2026 · billed at the s1-pro list rate
€0.070
€0.074
reax-1.3 REAX 1.3 by Decision Models · 1 question per request · 28,672 billed input tokens · EU only · output free · Model page
not offered
€0.400
Rate limits: 1,200 requests per minute per key by default, higher on request. See Errors and retries.
Requests that contain an image are billed at the image rate for all their input tokens (text and image positions), because image processing takes about 3× the GPU time per token. Text-only s1-vision requests use the s1-vision text rate. The old model ids s1-pro-1, s1-fast-1 and s1-vision-1 are retired and answered by the models above, at the same prices. On keys with peer-to-peer network processing enabled, s1-fast-1 may instead be answered by its original model (listed in the model licences); model_version shows which model answered.
The rate schedule and conversion basis are summarized in launch pricing sources; earlier schedules stay available as 3 October 2026. Peer-to-Peer first uses spare EU capacity, with EU requests taking priority. When busy, Peer-to-Peer can draw on rented worldwide GPU capacity from independent hardware providers in various countries, including outside the EU/EEA. This happens only for keys with “Allow worldwide processing” enabled. Without opt-in, Peer-to-Peer stays on EU capacity and may be queued or rejected when busy. Existing customers keep EU-only processing unless they opt in. Our API gateway does not log request or response content; on keys with the separate network-processing opt-in, independent providers of the peer-to-peer network may store content. Requests to models served by the network (currently: s1-fast-1) go to REAX miners with a track record as the primary route, with our GPUs as fallback; answers are checked on a sampled basis. All other models in this tier run on our GPUs (EU, and with worldwide processing enabled also non-EU partner GPUs). If the network has no capacity for an exact network version, the request is answered on our own GPUs with the same exact model version, never with another model. A retryable 503 is returned only if neither the network nor our own GPUs have capacity. This applies to keys with network processing turned on, which is preselected when you create a Peer-to-Peer key in the dashboard ("Our GPUs only" is selectable) and separate from worldwide processing; existing keys are not changed. The EU tier never uses the network. Rates use each request’s actual model input-token count.
Two approaches
Choose the approach that fits your data.
Same model profiles and API request shape. Implementations depend on available capacity. Choose per API key or per request with one header.
PEER-TO-PEERElastic capacity
Peer-to-Peer — capacity that scales with your workload
Built to scale worldwide. Peer-to-Peer starts on spare EU capacity. When it is busy, keys you opted in can draw on rented worldwide GPU capacity from independent hardware providers in various countries, including outside the EU/EEA.
Today
EU requests have priority. Enable “Allow worldwide processing” for each key that may leave the EU. Without opt-in, Peer-to-Peer stays on EU capacity and may be queued or rejected when busy. Peer-to-Peer capacity depends on how many network providers are live, so short bursts can get a retryable 503; for steady high volume use an EU key. Existing customers keep EU-only processing unless they opt in. Our gateway does not log payloads; on keys with the separate network-processing opt-in, independent providers of the peer-to-peer network may store content.
EU — for the strictest data-protection requirements
For teams with the highest data-protection demands or strict EU regulatory requirements. EU requests run only on contracted operators inside the EU, and inference data never leaves the EU. EU requests have priority, even if a key allows worldwide Peer-to-Peer processing.
Sign-in (Google) and payments (Stripe) are handled by the sub-processors listed on our sub-processor page; they never receive your prompts. Worldwide Peer-to-Peer capacity is currently rented through Lium (lium.io), whose independent GPU providers are listed there too. Network processing requires its own per-key opt-in; independent providers may store content.
Workload calculator
Estimate monthly usage
Choose a model rate, then enter your estimated volume. This is a cost estimate, not an invoice.
Model tokenizers can count the same text differently. Tax and other fees are not included.
Estimated monthly cost
—
Enter your workload to calculate an estimate.
Price changeEnter workload
Assumes 30 days per month and the selected rate. For s1-pro the estimate also shows the monthly cost at the rate existing customers keep until 12 November 2026. Estimates only; your invoice follows actual input tokens.
ACTUAL MODEL USAGE
Input tokens
Usage is based on the selected model’s actual token count. s1-pro accepts up to three text questions; s1-fast and s1-vision accept one. The limit depends on the model (8,192 to 28,672 input tokens for each question, including its state; table). Read the request limits →.
NO OUTPUT CHARGE
Output tokens
The API returns decisions rather than generated prose. Output tokens are not billed.