Token Pricing
Model and service pricing for EU and US inference, including input, cached input, output tokens, and speech-to-text.
Prices are in EUR. Choose an inference region to see prices for models available there. EU is selected by default. Set your API base URL as described in Region Inference to choose where requests run.
Model pricing
EU endpoint (default): https://api.greenpt.ai/v1
Chat and coding models
| Model | Input / 1M | Cached input / 1M | Output / 1M |
|---|---|---|---|
kimi-k3 | €3.30 | €0.825 | €16.50 |
glm-5.3 | €1.10 | €0.275 | €4.40 |
glm-5.3-flash | €0.11 | €0.022 | €0.44 |
deepseek-v4.1-flash | €0.22 | €0.011 | €1.10 |
deepseek-v4.1-flash-1017 | €0.36 | €0.04 | €1.45 |
minimax-m3 | €0.44 | €0.11 | €2.20 |
Embedding and reranking
These services charge only for input tokens, with no output token costs.
| Model | Service | Input / 1M |
|---|---|---|
green-embedding | Embedding | €0.20 |
qwen3-embedding-8b | Embedding | €0.25 |
green-rerank | Reranking | €0.12 |
Speech-to-text
Audio transcription services, with rates quoted per hour. Billing is based on audio duration.
| Model | Mode | Rate |
|---|---|---|
green-s | Pre-recorded audio (batch) | €0.23 / hour |
green-s | Live audio (real-time) | €0.31 / hour |
green-s-pro | Monolingual: pre-recorded | €0.23 / hour |
green-s-pro | Monolingual: live audio | €0.31 / hour |
green-s-pro | Multilingual: pre-recorded | €0.28 / hour |
green-s-pro | Multilingual: live audio | €0.37 / hour |
Additional features
Launch promo: the speech-to-text add-ons below are free right now. The prices shown are the standard per-hour rates that apply once the promo ends; you are not charged for them today.
| Feature | Rate (promo: free now) |
|---|---|
| Redaction | €0.10 / hour |
| Entity detection | €0.08 / hour |
| Streaming diarization (live audio) | €0.10 / hour |
| Keyterm prompting | €0.07 / hour |
Speaker diarization is included free for pre-recorded audio. Multichannel audio is billed per channel, so 2-channel audio is charged at double the per-hour rate.
Note: embedding, reranking, and speech-to-text services don't generate output tokens. They're billed only on input.
US endpoint: https://api.us.greenpt.ai/v1
Chat and coding models
| Model | Input / 1M | Cached input / 1M | Output / 1M |
|---|---|---|---|
kimi-k3 | €3.30 | €0.33 | €16.50 |
glm-5.3 | €1.595 | €0.1595 | €4.95 |
glm-5.3-flash | €0.165 | €0.033 | €0.55 |
deepseek-v4.1-flash | €0.165 | €0.0165 | €0.66 |
Embeddings
Embedding models charge only for input tokens, with no output token costs.
| Model | Service | Input / 1M |
|---|---|---|
qwen3-embedding-8b | Embedding | €0.011 |
Speech-to-text and reranking are currently available in EU inference only.
These tables show the active model catalog. For capabilities, not-recommended and deprecated IDs, and the dated history of price and status changes, see Model Cards.
Prompt caching
The Cached input column is what a prompt-cache hit costs on that model. Repeated prompt prefixes are served from cache automatically and billed at that lower rate; a dash means the model has no cache discount and bills every input token at the full rate. See Prompt caching. Output tokens are never discounted by prompt caching.
Token usage examples
A request is billed on the tokens it sends plus the tokens it generates, so the cost of a call follows from the rates above. As a rule of thumb, 1,000 tokens is roughly 750 words of English text. A few worked examples:
| Example | Model | Input tokens | Output tokens | Cost |
|---|---|---|---|---|
| A short question and answer | glm-5.3-flash | 300 | 150 | €0.000099 |
| A page of text summarized | deepseek-v4.1-flash | 8,000 | 500 | €0.00231 |
| An analysis with reasoning | kimi-k3 | 2,000 | 800 | €0.0198 |
| A coding turn on a warm cache | glm-5.3 | 20,000 (15,000 cached) | 1,200 | €0.014905 |
| A full repository file read into the prompt | glm-5.3 | 100,000 | 2,000 | €0.1188 |
Two details the examples above show:
- Reasoning tokens are output tokens. What the model generates while
thinking is included in
completion_tokensand billed at the output rate. - Cached input is cheaper. In the coding example, 15,000 of the 20,000 input
tokens are a repeated prompt prefix, billed at
glm-5.3's cached rate of €0.275 instead of €1.10. See Prompt caching.
EU service examples for services billed on input or audio duration: embedding 500,000 tokens with Green Embedding costs €0.10, reranking 200,000 tokens with Green Rerank costs €0.024, and transcribing an hour of pre-recorded audio with GreenS costs €0.23.
Pricing notes
- All prices are in EUR and exclude applicable taxes.
- Token counting follows OpenAI-compatible standards.
- On models with a cached-input rate, repeated prompt prefixes are billed at
that rate instead of the full input rate. It applies automatically, and the
response reports the count on
usage.prompt_tokens_details.cached_tokens. See Prompt caching. - Speech-to-text pricing is based on audio duration, not tokens.
- Output tokens are typically priced higher than input tokens.
- Embedding and reranking models only charge for input tokens (no output costs).
- Volume discounts available for enterprise customers.
- Prices subject to change with 30 days notice.