Poolside Laguna API Pricing Calculator
Estimate your monthly and annual Poolside Laguna API costs based on token volume, model tier, prompt caching, and workload type. Built for developers and engineering teams.
Estimate Your API Cost
Configure your expected monthly usage below
Best for everyday coding tasks, code review, and documentation.
Input = your prompts, system messages, context, and code you send to the API.
Output = code, explanations, or text generated by Laguna in response.
Reuses repeated context (system prompts, large codebases) at a lower rate. Highly effective for RAG and agent workloads.
Portion of your input tokens that are served from cache. These are billed at the cache read rate.
Helps estimate per-request cost breakdown.
Cost Estimate
Monthly & annual Poolside Laguna API spend
Configure your expected token usage, select a model, and click Calculate to see your Poolside Laguna API cost estimate.
Poolside Laguna Model Rates
Indicative per-token pricing for Poolside Laguna API tiers. All prices in USD per million tokens (MTok).
| Model | Input (per MTok) | Output (per MTok) | Cache Read (per MTok) | Best For |
|---|---|---|---|---|
| Laguna Standard | $1.00 | $3.00 | $0.10 | Lightweight tasks, high-volume automation |
| Laguna Plus | $3.00 | $15.00 | $0.30 | Everyday engineering, code review, RAG |
| Laguna Pro | $8.00 | $40.00 | $0.80 | Complex reasoning, large codebase analysis |
* Rates are indicative estimates for planning purposes. Always refer to the official Poolside documentation for current pricing.
Poolside Laguna API FAQ
Everything developers need to know about Poolside Laguna API costs and token usage.
Poolside Laguna is a large language model developed by Poolside AI, specifically optimised for software engineering tasks including code generation, debugging, refactoring, and technical documentation. It is available via API for developers and enterprises building AI-powered development tools.
Costs are based on the number of tokens processed — both input tokens (your prompts and context) and output tokens (the model’s generated response). Rates vary by model tier and are billed per million tokens (MTok). Output tokens are typically priced higher than input tokens.
A token is roughly 3–4 characters of text or about 0.75 words. For code, tokens tend to be shorter due to special characters and syntax. A typical 500-word engineering prompt uses around 670 tokens. One million tokens is roughly 750,000 words — or a large codebase context.
Prompt caching lets you store repeated context — like system prompts, large documents, or codebases — so they are not re-processed every request. Cached input tokens are billed at a cache read rate, typically around 90% cheaper than the full input rate. For RAG pipelines or agent workflows with consistent system prompts, caching can reduce costs significantly.
Laguna Standard is ideal for high-volume, lightweight tasks like unit test generation or simple completions. Laguna Plus suits most engineering workloads — code review, documentation, and RAG. Laguna Pro is best for complex multi-file reasoning, large codebase analysis, and architectural tasks where output quality is critical.
Key optimisations include: 1) Enable prompt caching for repeated context. 2) Keep system prompts concise. 3) Set max_tokens to limit unnecessarily long responses. 4) Route simple tasks to Laguna Standard. 5) Batch similar requests where the API supports it. 6) Strip irrelevant code context before sending it as input.
Poolside typically offers a free trial credit for new API users to evaluate the model before committing to production usage. Trial limits and terms vary — check the official Poolside documentation for the latest information on trial access and enterprise agreements.