Pay as you go
Model ratesStart with per-model input and output token pricing.
- No large upfront contract
- Estimate before launch
- Use OpenAI-compatible requests
/ pricing
Pay for what
youuse.
If you're comparing models for a live product, use the calculator first, then choose the plan that fits your spend pattern.
Credits
Prepaid balanceKeep spend predictable for experiments and small teams.
- Project-level budgeting
- Usage export friendly
- Good for pilots and QA
Team
Governed usageManage projects, keys, budgets, and model policy for production teams.
- Spend by project and key
- Budget controls
- Team reporting
BYOK
Use your keysBring existing provider accounts into one comparison and governance layer.
- Provider keys under your control
- Unified model catalog
- Policy-ready routing
Enterprise
CustomPrivate commercial terms for high-volume or governed workloads.
- Volume planning
- Provider mix design
- Support and rollout review
Calculator
Model, input tokens, output tokens, requests, and monthly estimate.
Use this as a pre-production estimate. Final billing should be reconciled against provider usage and platform usage records.
Estimate formula
Cost = requests x ((input tokens x input price) + (output tokens x output price)) / 1,000,000.
The default Doubao Seed 2.0 Mini estimate for 1M input and 1M output is $0.318.
Price calculator
Estimate monthly spend from model price, tokens, and request volume.
Monthly estimate$0.26Doubao Seed 2.0 Mini · input tokens 0.02893518518518519/1M · output tokens 0.28935185185185186/1M
How do you estimate AI API cost?
AI API cost is estimated by multiplying request count by input tokens and output tokens, then applying each model's posted price per 1M tokens. Teams should calculate a low-cost model, a quality fallback, and an expected monthly request volume before routing production traffic.
Want to estimate your API savings first?
Run CacheSafety Bench before enabling a cache policy in production. Bad Hit Rate matters more than raw hit rate.
Run CacheSafety BenchLow-cost reference
Compare low-cost model candidates before committing traffic.
Price is only one dimension. Review context length, capabilities, source labels, and intended use cases before production use.
| Model | Provider | Input | Output | Context | Capabilities | Best for | Latency | Status | Source |
|---|---|---|---|---|---|---|---|---|---|
| Doubao Seed 2.0 Minidoubao-seed-2-0-mini | Volcengine | $0.029 / 1M tokens | $0.289 / 1M tokens | 128k | Tool callingVisionJSON modeLong context | Chinese Q&A, low-cost general chat | 900-2600ms | Production | Platform curated |
| DeepSeek V4 Flashdeepseek-v4-flash | DeepSeek | $0.112 / 1M tokens | $0.224 / 1M tokens | 128k | Tool callingJSON modeLong contextReasoning | low-cost Chinese tasks, long-context summary | 700-2200ms | Catalog | OpenRouter if available |
| Mistral: Mistral Small 3.2 24Bmistralai/mistral-small-3.2-24b-instruct | Mistral AI | $0.1 / 1M tokens | $0.3 / 1M tokens | 128k | Tool callingJSON modeStreamingLow cost | translation, classification | 700-2300ms | Catalog | OpenRouter if available |
| DeepSeek V4 Flashdeepseek-v4-flash | DeepSeek | $0.145 / 1M tokens | $0.289 / 1M tokens | 128k | Tool callingJSON modeStreamingLow cost | high-volume chat, lightweight agent steps | 700-2200ms | Production | Platform curated |
| Doubao Seed 2.0 Litedoubao-seed-2-0-lite | Volcengine | $0.087 / 1M tokens | $0.521 / 1M tokens | 256k | Tool callingJSON modeLong contextStreaming | high-volume chat, classification | 800-2600ms | Production | Platform curated |
| OpenAI: GPT-4o-miniopenai/gpt-4o-mini | OpenRouter | $0.15 / 1M tokens | $0.6 / 1M tokens | 128k | Tool callingVisionJSON modeLong context | low-cost chat, image understanding | 800-2400ms | Catalog | OpenRouter if available |
| Meta: Llama 4 Maverickmeta-llama/llama-4-maverick | Meta | $0.15 / 1M tokens | $0.6 / 1M tokens | 1M | JSON modeLong contextStreamingLow cost | open-model workflows, cost-sensitive long context | 950-2800ms | Catalog | OpenRouter if available |
| MiniMax M2.7minimax-m2-7 | MiniMax | $0.243 / 1M tokens | $0.975 / 1M tokens | 128k | Tool callingJSON modeStreamingLow cost | high-volume chat, agentic tool use | 900-2800ms | Production | Platform curated |
FAQ
Pricing questions for product and platform teams
How is cost estimated?
The calculator multiplies input tokens and output tokens by the selected model price per 1M tokens, then multiplies by request count.
Does Doubao Seed 2.0 Mini still price at $0.318 for 1M input plus 1M output?
Yes. $0.029 input plus $0.289 output equals $0.318 for that single 1M + 1M request estimate.
Can teams use existing provider keys?
Yes. The BYOK plan is designed for teams that already have provider accounts and want consistent policy and usage reporting.
Is enterprise pricing available?
Enterprise pricing can be negotiated around volume, provider mix, region, support requirements, and governance needs.
How does LLM API pricing actually work?
LLM API pricing is per token, billed separately for input and output at a rate per 1M tokens. Your effective bill also depends on average output length, retries, streaming, and cache hit rate, so compare models on real prompts rather than the sticker rate alone.
How do I compare AI API pricing across providers?
Line up input price, output price, context window, and capability for each model, then estimate monthly cost from your token volume with the calculator at /tools/llm-cost-calculator before routing production traffic.