Pricing
Pricing for teams at every stage
Start free and pay per token, then talk with our team when you need a setup tuned to your traffic.
Pay as you go
$0+ usage
Everything you need to try open models and ship your first project.
What's included
- No commitment
- Free starter credits
- Per-token billing from prepaid credits
- OpenAI + Anthropic SDK compatible
- Lower rate on repeated input
Tailored
Custommonthly minimum
A setup tuned to your traffic, for teams running production workloads.
What's included
- Discounted rates for committed spend
- 2-week free evaluation on your traffic
- Tuned model and serving setup
- Dedicated capacity
- SLA on uptime, throughput, and latency
- Direct engineer support
Per-token rates
Every model, priced per million tokens
Pay-as-you-go rates in USD. Repeated input, such as a system prompt sent with every request, is billed at the cached rate.
| Model | Context | Input | Cached input | Output |
|---|---|---|---|---|
| DeepSeek V4 FlashDeepSeek | 1M | $0.04 | $0.02 | $0.21 |
| MiMo-V2.6 Flash | 1M | $0.14 | $0.003 | $0.28 |
| Qwen3.6-35B-A3BQwen | 262K | $0.04 | $0.02 | $0.35 |
| GLM-5.3 Flash | 1M | $0.12 | $0.02 | $0.40 |
| GPT-6 Luna | 1M | $0.10 | $0.01 | $0.50 |
| Nemotron 3 Super | 262K | $0.12 | $0.06 | $0.52 |
| MiniMax M3.1 FlashMiniMax | 1M | $0.15 | $0.02 | $0.60 |
| Gemma 4 31B | 262K | $0.42 | $0.10 | $0.84 |
| MiMo-V2.6 Pro | 1M | $0.44 | $0.004 | $0.87 |
| DeepSeek V4.1 Flash 1017DeepSeek | 1M | $0.24 | $0.005 | $0.96 |
| DeepSeek V4.1 FlashDeepSeek | 1M | $0.20 | $0.01 | $1.00 |
| MiniMax M3MiniMax | 1M | $0.28 | $0.06 | $1.10 |
| GPT-5.6 Luna | 262K | $0.20 | $0.02 | $1.20 |
| DeepSeek V4 ProDeepSeek | 1M | $1.04 | $0.04 | $2.08 |
| GLM-5.2 | 1M | $0.34 | $0.17 | $2.30 |
| GLM-5.3 | 1M | $1.12 | $0.14 | $3.52 |
| GPT-5.6 Sol | 262K | $2.00 | $0.20 | $10.00 |
| GPT-6 Sol | 1M | $2.00 | $0.20 | $10.00 |
| Kimi K3 | 1M | $2.30 | $0.23 | $11.40 |
| Claude Opus 5.5 | 1M | $4.00 | $0.20 | $20.00 |
| Claude Opus 5 | 262K | $5.00 | $0.50 | $25.00 |
| Claude Fable 5.1 | 262K | $10.00 | $0.25 | $50.00 |
FAQ
Pricing questions
How much does Boundless cost?
Pay as you go has no fee and no commitment: you pay per token at the rates listed for each model. Tailored has a monthly minimum spend agreed with our team, with discounted rates for committed spend.
How does pay-as-you-go billing work?
You pay per token from prepaid credits. Requests stop when credits run out, so there are no surprise bills, and you can set a budget for each API key.
How is Tailored priced?
Tailored has a monthly minimum spend that we agree with your team, and committing to more lowers your per-token rate. We set the rate after the evaluation, once we know the model, setup, and capacity your traffic needs.
What does the free evaluation include?
We agree on targets with your team, test your real traffic against your current provider for two weeks, and you decide from the results.
Can we start on pay as you go and move to Tailored later?
Yes. It's the same API, so moving needs no code changes, and we move your traffic once the evaluation beats your current setup.
What counts as repeated input?
Any part of a prompt we've already seen, such as a system prompt or shared context, is billed at the lower cached rate listed for each model.
Where does our data go?
We keep zero data retention on the open models we serve, so prompts and outputs aren't stored, and your requests can run in the US or Europe.